News

Preferred Networks raises a total of over 2 billion yen from FANUC, Hakuhodo DYHD, Hitachi, Mizuho Bank, and Mitsui & Co.

December 11, 2017, Tokyo Japan – Preferred Networks, Inc. (PFN, Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toru Nishikawa) has agreed to allocate new shares in December 2017 to Hakuhodo DY Holdings Inc. (Headquarters: Minato-ku, Tokyo, President and CEO: Hirokazu Toda), Hitachi, Ltd. (Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toshiaki Higashihara), Mizuho Bank, Ltd. (Headquarters: Chiyoda-ku, Tokyo, President and CEO: Koji Fujiwara) and Mitsui & Co., LTD. (Headquarters: Chiyoda-ku, Tokyo, President and CEO: Tatsuo Yasunaga). PFN will receive an investment of about 500 million yen from each of them.

Similarly, FANUC CORPORATION (FANUC, Headquarters: Oshino-mura, Yamanashi Prefecture, Chairman and CEO: Yoshiharu Inaba) has also agreed to acquire additional shares of PFN for about 500 million yen. FANUC and PFN formed an R&D alliance in June 2015 and a capital alliance in August of the same year.

With these deals, PFN will complete a series of fund-raising activities since the additional investment of about 10.5 billion yen from Toyota Motor Corporation, which was announced in August 2017.

PFN will use the raised capital to strengthen its financial base, improve computing resources, and hire talented people while collaborating with each of the investing companies in respective fields for growth over the mid- to long-term.

PFN will strive to drive innovation not only in the fields of transportation system, manufacturing, and bio/healthcare but in a broad range of business areas to increase its corporate value.

 

We received comments from the companies investing in PFN.

 “FANUC will apply Preferred Networks’ cutting-edge machine-learning and deep-learning techniques to our FIELD system* and all the other products to provide the most effective production system that processes information quickly and in real time at the edges of machining and assembly lines in production sites, enabling machines to collaborate with each other in a flexible and smart way.”

Yoshiharu Inaba
Chairman and CEO of FANUC

 

*FANUC Intelligent Edge Link and Drive system is an open platform to optimize manufacturing by interconnecting various devices in production sites and is available for a variety of companies to join.

 

“I have high expectations that we can create innovative solutions and services by combining Preferred Networks’ deep learning and other machine-learning techniques and implementation capabilities, and Hakuhodo DY Group’s proprietary data and creativity in the fields of advertising, marketing, media, and content. We will, in collaboration with Preferred Networks, promote the fusion of creativity and technology for implementation in the real world and realize innovations that make life richer and more enjoyable.”

Hirokazu Toda
President and CEO of Hakuhodo DY Holdings

 

“Hitachi is utilizing leading-edge technologies, including IoT and AI, and enhancing collaborative creation with various partners in the Social Innovation Business. Taking the investment as an opportunity to combine the strengths of both companies such as Preferred Networks’ knowledge of deep learning and Hitachi’s technology and know-how in the integration of OT, IT, and products, we will begin collaborative creation towards the realization of Society 5.0.”

Yutaka Saito
Representative Executive Officer, Executive Vice President and Executive Officer of Hitachi

 

“Preferred Networks is a leading company in the industry that has a high level of AI technology. Through the new partnership with Preferred Networks, Mitsui will not only promote higher operation efficiency and higher value added of our global business assets but also accelerate Digital Transformation activities that lead to the creation of new businesses with the aim of strengthening competitiveness.”

Nobuaki Kitamori
Representative Director, Executive Managing Officer, and Chief Digital Officer of Mitsui & Co.

 

Related link:

Hitachi Invests in Preferred Networks
http://www.hitachi.com/New/cnews/month/2017/12/171211.html

Mitsui to Invest in AI Company Preferred Networks, Inc.
https://www.mitsui.com/jp/en/release/2017/1225186_10832.html

 

 

 

◆ About Preferred Networks, Inc.

Founded in March 2014 with the aim of promoting business utilization of deep learning technology focused on IoT, PFN advocates Edge Heavy Computing as a way to handle the enormous amounts of data generated by devices in a distributed and collaborative manner at the edge of the network, driving innovation in various fields, with a focus on three business areas: transportation, manufacturing, and bio/healthcare. PFN develops and provides Chainer®, an open source deep learning framework. PFN promotes advanced initiatives by collaborating with world leading organizations, such as Toyota Motor Corporation, FANUC CORPORATION, and the National Cancer Center. (https://www.preferred-networks.jp/en/)

 

*Company names and product names written in this release are the trademarks or the registered trademarks of each company.

Preferred Networks’ private supercomputer ranked first in the Japanese industrial supercomputers TOP 500 list

MN-1, a private supercomputer used exclusively by Preferred Networks, Inc. (PFN, Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toru Nishikawa) has recorded a LINPACK  performance of about 1.39 PetaFLOPS 2.  As a result, MN-1 is ranked 12th in the world and 1st in Japan among industrial supercomputers in the TOP500 List (http://www.top500.org), which shows the most powerful supercomputers as of November 2017. When including supercomputers for research purposes, it is listed as 91st in the world and 13th in Japan.   

 

 

About PFN’s private supercomputer MN-1   

MN-1 is equipped with a Graphics Processing Unit (GPU) platform from NTT Communications Corporation (NTT Com, Headquarters: Chiyoda-ku, Tokyo, President and CEO: Tetsuya Shoji) and NTT PC Communications Incorporated (NTTPC, Headquarters: Minato-ku, Tokyo, President and CEO: Motoo Tanaka), and contains 1,024 of NVIDIA (R) ’s Tesla (R) multi-node P100 GPUs. It utilizes Mellanox’s Infiniband interconnect to make high-speed distributed deep learning possible using ChainerMN 4, a distributed deep learning package developed by PFN.    

Using MN-1, PFN will further accelerate its research and development activities in various fields that require a huge amount of computing resources such as transportation systems, manufacturing and bio/healthcare.

 

 

 

1.  A benchmark to compare practical operation speed of computers

2.  A unit measuring computer performance. Peta is 1,000 trillion (10 to the power of 15) and FLOPS is used to count floating-point operations per second. So, 1 PetaFLOPS means that a computer is capable of performing 1,000 trillion floating-point calculations per second.

3.  Preferred Networks launches one of Japan’s most powerful private sector supercomputers
https://www.preferred-networks.jp/en/news/pr20170920

4.  Preferred Networks achieved the world’s fastest training time in deep learning
https://www.preferred-networks.jp/en/news/pr20171110

 

About Preferred Networks, Inc.

Founded in March 2014 with the aim of promoting business utilization of deep learning technology focused on IoT, PFN advocates Edge Heavy Computing as a way to handle the enormous amounts of data generated by devices in a distributed and collaborative manner at the edge of the network, driving innovation in three priority business areas: transportation, manufacturing and bio/healthcare. PFN promotes advanced initiatives by collaborating with world leading organizations, such as Toyota Motor Corporation, FANUC CORPORATION, and the National Cancer Center. (https://www.preferred-networks.jp/en/)

 

*Chainer (R)  is the trademark or the registered trademark of Preferred Networks, Inc. in Japan and other countries. Other company names and product names written in this release are the trademarks or the registered trademarks of each company.

Preferred Networks achieved the world’s fastest training time in deep learning, completed training on ImageNet in 15 minutes,using the distributed learning package ChainerMN and a large-scale parallel computer

November 10, 2017, Tokyo – Preferred Networks, Inc. (PFN, Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toru Nishikawa) has achieved the world’s fastest training time in deep learning by using its large-scale parallel computer MN-1 1.

With the size of training data and the number of parameters expanding for the sake of better accuracy of deep learning models, computation time is also increasing. It is not unusual to take several weeks to train a model. Getting multiple GPUs to link with one another for faster training is very important to reduce the time spent on trial and error and verification of new ideas, and produce results quickly.

On the other hand, it is generally known in parallel/distributed learning that the accuracy and learning rate of a model decrease gradually with increased GPUs, due to larger batch sizes and GPU communication overhead.

This time, we have improved learning algorithms and parallel performance to address these issues, and used one of Japan’s most powerful parallel computers with 1,024 of NVIDIA(R)’s Tesla(R) multi-node P100 GPUs and leverages Chainer’s distributed learning package ChainerMN 2 for training.

As a result, we completed training ResNet-50 3 for image classification on the ImageNet 4 dataset in 15 minutes, which is a significant improvement from the previously best known result 5.

The research paper on this achievement is available in the following URL under the title “Extremely Large Minibatch SGD: Training ResNet-50 on ImageNet in 15 Minutes”. (https://www.preferred-networks.jp/docs/imagenet_in_15min.pdf)

Based on this research result, PFN will further accelerate its research and development activities in the fields of transportation systems, manufacturing, and bio/healthcare, which require large-scale deep learning.

 

1 One of the most powerful private supercomputer in Japan, contains 1,024 of NVIDIA(R)’s Tesla(R) multi-node P100 GPUs.https://www.preferred-networks.jp/en/news/pr20170920

2 A package adding distributed learning functionality with multiple GPUs to the open source deep learning framework Chainer

3 A network frequently used in the field of image recognition

4 A dataset widely used for image classification

5 Training completed in 31 minutes using Intel(R) Xeon(R) Platinum 8160 x 1,600(Y. You et al. ImageNet Training in Minutes. CoRR,abs/1709.05011, 2017)

 

■ About the Chainer Open Source Deep Learning Framework

Chainer is a Python-based deep learning framework being developed mainly by PFN, which has unique features and powerful performance that allow for designing complex neural networks easily and intuitively, thanks to its “Define-by-Run” approach. Since it was open-sourced in June 2015, as one of the most popular frameworks, Chainer has attracted not only the academic community but also many industrial users who need a flexible framework to harness the power of deep learning in their research and real-world applications.

Chainer incorporates the results of the latest deep learning research. With additional packages such as ChainerMN (distributed learning), ChainerRL (reinforcement learning), ChainerCV (computer vision) and through the support of Chainer development partner companies, PFN aims to promote the most advanced research and development activities of researchers and practitioners in each field. (http://chainer.org/

■ About Preferred Networks, Inc.

Founded in March 2014 with the aim of promoting business utilization of deep learning technology focused on IoT, PFN advocates Edge Heavy Computing as a way to handle the enormous amounts of data generated by devices in a distributed and collaborative manner at the edge of the network, driving innovation in three priority business areas: transportation, manufacturing and bio/healthcare. PFN promotes advanced initiatives by collaborating with world leading organizations, such as Toyota Motor Corporation, Fanuc Corporation, and the National Cancer Center. (https://www.preferred-networks.jp/en/)

*Chainer(R) is the trademark or the registered trademark of Preferred Networks, Inc. in Japan and other countries.

 

 

AI (Machine Learning) Improves Wire-cut EDM Accuracy

FANUC CORPORATION (hereinafter, FANUC) in collaboration with Preferred Networks, Inc. (hereinafter, PFN) has developed an AI thermal displacement compensation function which will improve the machining accuracy of ROBOCUT α-CiB series, FANUC’s wire-cut electric discharge machine (see Note 1 below). ROBOCUT with this function will be the first product using AI since FANUC and PFN began collaborating.

 

FANUC and PFN formed an R&D alliance*1 in June 2015, followed by a capital alliance*2 in August of the same year to promote a joint development of AI functions for the manufacturing industry that can efficiently improve the performance and operation rates of FANUC products. The newly developed function utilizes machine-learning (ML) technology to predict and control the variable machining accuracy caused by ROBOCUT’s temperature fluctuations, with 30% more accurate compensation than existing method. The new function is applicable from small to large workpieces.

The AI thermal displacement compensation function will be provided as an optional function to ROBOCUT, and FANUC plans to start accepting orders in November of this year. FANUC will also display the ROBOCUT with this new function at Mechatronics Technology Japan, which will be held in Port Messe Nagoya on Oct. 18-21, 2017.

ROBOCUT α-CiB series

FANUC is also developing a similar function for the ROBODRILL series that utilizes ML and expect to release it in the near future.
FANUC and PFN will continue making gradual but steady progress towards realizing innovative manufacturing fields through AI.

“It is my pleasure to announce our first product based on the machine-learning technology since the tie-up with FANUC. Through this product, we can demonstrate using ML is effective in optimizing control parameters, which is one of the most important issues facing the manufacturing industry. PFN will continue to contribute to the intelligence of machine tools and robots by applying machine learning and deep learning techniques.”
Toru Nishikawa, Chief Executive Officer of PFN

 

*1 Announcement for R&D alliance with FANUC Corporation
https://www.preferred-networks.jp/en/news/8731
*2 Announcement for capital tie-up between FANUC and PFN
http://www.fanuc.co.jp/en/profile/pr/newsrelease/notice20150821.html

 

Note 1. Wire-cut EDM is a precision and fine shape machining tool that uses discharge phenomenon between the ultrathin wire electrode and the metal workpiece (electric conductor).

 

Preferred Networks released open source deep learning framework Chainer v3 and NVIDIA GPU array calculation library CuPy v2

Preferred Networks, Inc. (PFN, Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toru Nishikawa) has released Chainer v3, a major update of the open source deep learning framework Chainer(R), as well as NVIDIA(R) GPU array calculation library CuPy™ v2.

We release a major upgrade of Chainer every three months that quickly incorporates the results of the latest deep learning research. The newly released Chainer v3 will run without the need to change most of your code.

 

Main features of Chainer v3 and CuPy v2 include:

1.  Automatic differentiation of second and higher order derivatives

Chainer now supports automatic differentiation of second order and higher derivatives in many functions. This will enable users to easily implement deep learning methods that require second order differentiation as per equations written in papers.

 

2. Improved CuPy memory allocation

In many neural nets, memory efficiency when using GPUs will improve significantly, and reallocation of memory will be reduced in some cases, increasing speed.

 

3. Sparse matrix support has been added to CuPy

Large-scale graph analysis and natural language processing, which have previously been highly costly to implement on GPUs, can now be implemented more easily thanks to sparse matrix calculation being available on the GPU.

◆ Chainer ReleaseNote: https://github.com/chainer/chainer/releases/tag/v3.0.0

Chainer v3 has taken in a number of development results from external contributors as its previous versions did. PFN will continue working with supporting companies and the OSS community to promote the development and popularization of Chainer.

 

◆ About the Chainer Open Source Deep Learning Framework

Chainer is a Python-based deep learning framework developed by PFN, which has unique features and powerful performance that enables users to easily and intuitively design complex neural networks, thanks to its “Define-by-Run” approach. Since it was open-sourced in June 2015, as one of the most popular frameworks, Chainer has attracted not only the academic community but also many industrial users who need a flexible framework to harness the power of deep learning in their research and real-world applications.

Chainer incorporates the results of the latest deep learning research. With additional packages such as ChainerMN (distributed learning), ChainerRL (reinforcement learning), ChainerCV (computer vision) and through the support of Chainer development partner companies, PFN aims to promote the most advanced research and development activities of researchers and practitioners in each field. (http://chainer.org/)

*Chainer(R) and CuPyTM are the trademark or the registered trademark of Preferred Networks, Inc. in Japan and other countries.

Preferred Networks Launches one of Japan’s Most Powerful Private Sector Supercomputers

Features NTT Com Group’s Cloud-based GPU platform

TOKYO, JAPAN — This September, Preferred Networks, Inc. (PFN), a provider of IoT-centric deep learning systems, NTT Communications Corporation (NTT Com), the ICT solutions and international communications business within the NTT Group, and NTT Com subsidiary NTT PC Communications Incorporated (NTTPC) announced today the launch of a private supercomputer designed to facilitate research and development of deep learning, including autonomous driving and cancer diagnosis.

The new supercomputer is one of the most powerful to be developed by the private sector in Japan and is equipped with NTT Com and NTTPC’s Graphics Processing Unit (GPU) platform, and contains 1,024 of NVIDIA(R)’s Tesla(R) multi-node P100 GPUs. Theoretically, the processing speed of the new supercomputer can reach 4.7 PetaFLOPS—a massive 4,700 trillion floating point operations per second—the fastest levels of any computing environment in Japan.

Overview of the private supercomputer

 

PFN’s deep learning research demands an ultra high-speed, high capacity, state-of-the-art computing environment. Existing GPU platforms require massive electricity supplies, generate excessive heat and offer inadequate network speed. To address these issues, PFN adopted the NTT Com Group’s proven GPU platform, which boasts significantly advanced technology. They additionally leveraged the latest data center design, building a large-scale multi-node platform using ChainerMN, PFN’s technology that significantly accelerates the speed of deep learning by parallelizing calculations over multiple nodes.

NTT Com group has developed and released a multi-node GPU platform on Enterprise Cloud, Nexcenter(TM), a world-leading data center service, which incorporates the group’s extensive know-how in GPU performance maximization.

Following the supercomputer launch, PFN plans to increase the processing speed of its open source deep learning framework Chainer. They will further accelerate their research and development in the field of transportation systems, manufacturing, bio and healthcare industry which require a huge amount of computing resources. PFN will additionally consider the deployment of NVIDIA(R) Tesla(R) V100 GPUs, which are based on next-generation Volta GPU technology. NTT Com group will continue to support PFN’s research and the commercialization of their developed solutions with AI-related technologies and platforms.

“NVIDIA is excited to see the launch of Preferred Networks’ private supercomputer, built in partnership with NTT Com Group. Computing power is the source of competitive advantage for deep learning, the core technology of modern AI. We have high expectations that the new system will accelerate Preferred Networks’ business and contribute to Japan’s economic growth.”

Masataka Osaki
NVIDIA Japan Country Manager, Vice President of Corporate Sales

 

Related links:

Chainer
Enterprise Cloud
Nexcenter

 

◆ About Preferred Networks, Inc.
Founded in March 2014 with the aim of promoting business utilization of deep learning technology focused on IoT, PFN advocates Edge Heavy Computing as a way to handle the enormous amounts of data generated by devices in a distributed and collaborative manner at the edge of the network, driving innovation in three priority business areas: transportation, manufacturing and bio/healthcare. PFN develops and provides Chainer, an open source deep learning framework. PFN promotes advanced initiatives by collaborating with world leading organizations, such as Toyota Motor Corporation, Fanuc Corporation, and the National Cancer Center.
https://www.preferred-networks.jp/en/

◆ About NTT Communications Corporation
NTT Communications provides consultancy, architecture, security and cloud services to optimize the information and communications technology (ICT) environments of enterprises. These offerings are backed by the company’s worldwide infrastructure, including the leading global tier-1 IP network, the Arcstar Universal One™ VPN network reaching over 190 countries/regions, and 140 secure data centers worldwide. NTT Communications’ solutions leverage the global resources of NTT Group companies including Dimension Data, NTT DOCOMO and NTT DATA.
www.ntt.com | Twitter@NTT Com | Facebook@NTT Com | LinkedIn@NTT Com

◆ NTT PC Communications Incorporated
NTTPC Communications Incorporated (NTTPC), established in 1985 is a subsidiary of NTT Communications, is a network service and communication solution provider in Japanese telco market, the company has been the most strategic technology company of the group throughout of years. NTTPC launched the 1st ISP service of the NTT group, so called “InfoSphere” at 1995, and also launched the 1st Internet Data Center and server hosting services of Japan so called “WebARENA” at 1997. NTTPC have always started something new in ICT market.
http://www.nttpc.co.jp/english/

 

 

Notes
1. Chainer(R) is the trademark or the registered trademark of Preferred Networks, Inc. in Japan and other countries.

2. Other company names and product names written in this release are the trademarks or the registered trademarks of each company.

Call for application for PFN summer internship 2018 in Tokyo

Preferred Networks (PFN) will be organizing internship programs next summer in Tokyo. In order to make the process smooth for students from outside of Japan, we open an early bird application for them.

We are a growing startup with about 100 members based in Tokyo, Japan, focusing on applying deep learning to industrial problems such as autonomous driving, manufacturing, and bio-healthcare. We are actively developing the deep learning framework Chainer.

We look for brilliant students who have expertise on various topics, such as deep learning, reinforcement learning, computer vision, bioinformatics, natural language processing, distributed computing, simulation, etc.

In previous years, by selecting highly capable interns and encouraging them to tackle challenging and important problems, some of the internship results have been published at top conferences such as ICML or workshops at ICRA and ICCV.

During the internship, you will have unique opportunity to collaborate with highly motivated experts for working on real-world applications of deep learning, while staying in Tokyo, one of the most attractive cities in the world.

We are looking forward to receiving your applications, following the instructions below.

 

● Target of this program

  • Students outside of Japan

 

Work time & Location:

  • Business hours:
    8 hours/day, 5 days/week (excluding national holidays)
  • Location: Center of Tokyo
    Preferred Networks Tokyo office: Otemachi Bldg. 2F, 1-6-1, Otemachi, Chiyoda-ku, Tokyo, Japan 100-0004
    https://www.preferred-networks.jp/en/about

 

Period & Compensation:

  • The period of the internship can be flexibly arranged though it usually starts in June and finish by the end of August
  • We require minimum of two months (40 business days), in order to be able to tackle a challenging task
  • Interns are paid a competitive salary
  • We will cover residence and travel cost

 

Requirements:

  • Experience in at least one of the technology areas (listed below) other than lectures
    e.g., published a paper, won a competition, part-time work, open source contribution
  • Strong programming skill (any programming language)
  • Formally enrolled in university or research institute outside of Japan during 2018-2019 school year
  • Fluent in either English or Japanese
  • Able to work fulltime on weekdays at our Tokyo office during the period

 

Preferred experience & skills:

  • Machine learning and deep learning
  • Experience with numpy / scipy / deep learning frameworks
  • Experience with software & service development
  • Experience working with shared codebases (e.g. github / bitbucket / etc)
  • Contribution to open source projects

 

Candidate themes (subject to change)

1. Technology areas: Sub-field of machine learning, such as

     a. Deep learning theory

     b. Reinforcement learning

     c. Computer vision

     d. Natural language processing

     e. Parallel / distributed computing

 

2. Application areas: Advanced applications, such as

     a. Object detection / tracking / segmentation from image / video

     b. Robotics / factory automation / predictive maintenance

     c. Life science / healthcare / medicine

     d. Human machine interaction

     e. Design / content creation / visualization

     f. Deep learning software (Chainer, CuPy, ChainerMN/CV/RL, etc)

     g. Optimization for deep learning hardware

 

Application information:

  • Resume / CV (PDF format only. Please DO NOT include any personal or private information [e.g., age, race, nationality, religion, personal address, phone number] except name, email address, affiliation)
  • Github account (optional)

 

How to apply:

  • Please fill the google forms and submit
  • Due: September 29th, 11:59 pm Friday (PST)

  • No late submission will be accepted (we are planning to open the 2nd call for application by January, and another call for students in Japan by May)
  • The review process takes about 6-8 weeks after submission
  • Usually, getting a visa for working in Japan takes up to 3 months

 

Interview process:

  1. Document review
  2. One-way video interview (webcam, recording)
  3. Skype interview in English or Japanese (multiple times if necessary)

 

If you have questions, please contact us at hr-pfn@preferred.jp (Sorry but no late application is accepted for fairness)

Preferred Networks officially released ChainerMN version 1.0.0, a multi-node distributed learning package, making it even faster with stablized data-parallel core functions

Tokyo, Japan, September 1, 2017 – Preferred Networks, Inc. (PFN, Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toru Nishikawa) has released the official version 1.0.0 of ChainerMN※1, which is a package adding distributed learning functionality with multiple GPUs to Chainer, the open source deep learning framework developed by PFN.

For practical application of machine learning and deep learning technologies, the ever-increasing complexity of neural network models, with a large number of parameters and much larger training datasets requires more and more computational power to train these models.

ChainerMN is a multi-node extension to Chainer that realizes large-scale distributed deep learning by high-speed communications both intra- and inter-node. PFN released the beta version of ChainerMN on May 9, 2017 and this is the first official release. The following features have been added to ChainerMN v1.0.0.

● Features of ChainerMN v1.0.0

1. Increased stability in core functions during data parallelization

With this improved stability, ChainerMN can be used more comfortably.

2. Compatibility with NVIDIA Collective Communications Library (NCCL) 2.0.

By supporting the latest version, it has become even faster.

3. More sample code (machine translation, DCGAN) is available.

These examples will help users learn more advanced ways of using ChainerMN.

4. Expansion of supported environments (non-CUDA-Aware MPI).

CUDA – Aware MPI implementation such as Open MPI and MVAPICH was necessary for the beta version, but ChainerMN is now compatible with non-CUDA-Aware MPI.

5. Initial implementation of model parallelism functions.

More complex distributed learning has become possible by getting multiple GPUs to work in the model parallelism method.
The conventional data parallelism approach is known to limit the possible batch size when increasing the nodes while maintaining accuracy. o overcome this, we have done the initial part of the more challenging implementation of model parallelism for greater speed than possible with data parallelism.

 

These features provide a more stable and faster than ever deep learning experiences with ChainerMN, as well as improved usability.

The following is the result of the performance measurement of ChainerMN using the image classification dataset of ImageNet. It is about 1.4 times faster than the first announcement in January this year, and 1.1 times faster than the beta version released in May. Please visit the following Chainer Blog to learn more about the experiment settings:

https://chainer.org/general/2017/02/08/Performance-of-Distributed-Deep-Learning-Using-ChainerMN.html

 

In addition, from October 2017, ChainerMN will become available on “XTREME DNA”, an unmanned  cloud-based super-computer deployment and operation service, provided by XTREME Design Inc.(Head office: Shinagawa-ku, Tokyo, CEO: Naoki Shibata)

ChainerMN will be added on the distributed parallel environment templates for GPU instances of the pay-per-load public cloud, Microsoft Azure. This not only eliminates the need to construct infrastructure required for large-scale distributed deep learning but also makes it easy to manage research-and-development costs.

ChainerMN aims to provide an environment in which deep learning researchers and developers can easily concentrate on the main parts of research and development including the design of neural networks. PFN will continue to improve ChainerMN by adding more features and expanding its usage environment.

 

◆ The Open Source Deep Learning Framework Chainer (http://chainer.org)
Chainer is a Python-based deep learning framework developed by PFN, which has unique features and powerful performance that enables users to easily and intuitively design complex neural networks, thanks to its “Define-by-Run” approach. Since it was open-sourced in June 2015, as one of the most popular frameworks, Chainer has attracted not only the academic community but also many industrial users who need a flexible framework to harness the power of deep learning in their research and real-world applications.

 

※1:MN in ChainerMN stands for Multi-Node. https://github.com/pfnet/chainermn

Regarding the additional investment made by Toyota Motor Corporation announced on August 4.

Thank you for your inquiries concerning additional investment made by Toyota Motor Corporation announced on August 4.
We received a lot of questions about the shareholding ratio, so allow me to clarify.

The shareholding of Toyota Motor, which will be changed as a result of additional investment in PFN, will not change our existing business and management policy.

Preferred Networks received about 10.5 billion yen in investments from Toyota Motor Corporation

Preferred Networks received about 10.5 billion yen in investments from Toyota Motor Corporation. Accelerate joint research and development of AI technology in the mobility field

Tokyo, Japan, August 4, 2017 – Preferred Networks, Inc. (PFN, Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toru Nishikawa) agreed to receive an additional investment of approximately 10.5 billion yen from Toyota Motor Corporation (Toyota, Headquarters: Toyota  City, Aichi Prefecture, President: Akio Toyoda). PFN will allocate new shares to Toyota. After the allocation, Toyota will be the largest external shareholder of PFN.

PFN and Toyota started the joint research and development in October 2014, and Toyota invested 1.0 billion yen in PFN in December 2015 to strengthen the relationship.

Through the past joint research and development of object recognition technologies and vehicle information analysis, PFN’s world-leading intelligent technologies in machine learning, deep learning, and big data processing have been highly appreciated as essential for Toyota aiming at realizing the next generation of mobility society, including autonomous driving.

With this funding, PFN will enhance the computational environment and accelerate talent acquisition. We will further accelerate joint research and development by strengthening our relationship with Toyota in the field of mobility.

 

◆ About Preferred Networks, Inc. https://www.preferred-networks.jp/

Founded in March 2014 with the aim of business utilization of deep learning technology focused on IoT. PFN advocates Edge Heavy Computing as a way to handle the enormous amount of data generated by devices in a distributed and collaborative manner at the edge of the network and realizes innovation in the three priority business areas of the transportation system, manufacturing industry, and bio/healthcare.

PFN develops and provides Chainer, and open source deep learning framework, and solutions based on the Deep Intelligence in Motion (DIMo) platform that provides state-of-the-art deep learning technology. PFN promotes advanced initiatives by collaborating with world leading organizations, such as Toyota Motor Corporation, Fanuc Corporation, National Cancer  Center.

*Chainer® and DIMoTM are trademarks of Preferred Networks, Inc. in Japan and other countries.

 

Preferred Networks recieved the “Emerging Leader Award” at the 2017 Japan-U.S. Innovation Award

Preferred Networks recognized for pioneering application of Deep Learning Technology

    

 Receiving 2017 Emerging Leader Award (Japan) with Dr. Richard Dasher, Director, Stanford US-Asia Technology Management Center and Hiroshi Maruyama, Chief strategy officer of PFN.

Preferred Networks, Inc. was selected as the “2017 Emerging Leader Award”, the winner of the 2017 Japan-U.S. Innovation Award held at Stanford University on Friday, July 28th. PFN was one of the “Innovation ShowCase” start-up companies in 2016, and the rapid growth in the past year was highly appreciated.

US-Japan Innovation Awards is operated by the Japan Society of Northern California in collaboration with the Stanford University US–Asia Technology Management Center.

“Emerging Leader Awards” are presented to young, dynamic, and growing, entrepreneurial companies founded upon innovations that are demonstrating impressive success and that have the potential to make substantial global impacts. One U.S. and one Japanese award recipients are selected each year through a rigorous process guided by a distinguished 40-person Innovation Advisory Council composed of venture capitalists, academic experts, and prominent business executives based in Japan and in San Francisco/Silicon Valley.

Preferred Networks appoints Professor Pieter Abbeel of UC Berkeley as a Technical Advisor

Preferred Networks, Inc. (PFN, HQ: Chiyoda-ku, Tokyo, President & CEO: Toru Nishikawa) agreed with Pieter Abbeel (Professor at University of California, Berkeley, and Research Scientist at OpenAI) that he will be appointed as a technical advisor at PFN as of August 1st, 2017.

Professor Abbeel is a pioneer in machine learning based optimization and automation in robotics. He has published many research papers on robot control using deep reinforcement learning. Recently he joined OpenAI, a non-profit AI research company as research scientist.

In this appointment as a technical advisor, Professor Abbeel will give technical advice and guidance on cutting-edge deep learning to robotics research in PFN, aiming to accelerate the development and practical use of PFN’s technology. 

For more information, please refer to the research blog: https://preferredresearch.jp/2017/07/22/abbeel/

 

  • Pieter Abbeel

  • Biography

    He received a BS/MS in Electrical Engineering from KU Leuven (Belgium) and received his Ph.D. degree in Computer Science from Stanford University in 2008. He joined the faculty at UC Berkeley in Fall 2008, with an appointment in the Department of Electrical Engineering and Computer Sciences. His research focuses on robotics, machine learning and control. Professor Abbeel has won various awards, including IEEE Robotics and Automation Society Early Career Award, Sloan Research Fellowship. His work has been featured in many popular press outlets, including BBC, New York Times, and MIT Technical Review.

*DIMo™ are trademarks of Preferred Networks, Inc. in Japan and other countries.

Preferred Networks released Version 2 of Chainer, an Open Source framework for Deep Learning

New functions developed, including a significant increase of memory efficiency during learning

Tokyo, Japan, June 2, 2017 – Preferred Networks, Inc. (PFN, Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toru Nishikawa) has released a major update of its open source deep learning framework Chainer, called Chainer v2.

This is the first major version update since the official release of Chainer in 2015, and it enables more powerful, flexible, and intuitive functions to implement and study deep learning methods.

With the rapid evolution of deep learning technology and an expanding field of target applications, user demands for the functionality of deep learning frameworks is rapidly changing and diversifying.

Chainer incorporates the results of the latest deep learning research. With additional packages such as ChainerMN (distributed learning), ChainerRL (reinforcement learning), ChainerCV (computer vision) and through the support of Chainer development partner companies, PFN aims to promote the most advanced research and development activities of researchers and practitioners in each field.

 

Chainer v2 has three major enhancements and improvements.

 

1. Improved memory efficiency during learning

Chainer v2 shows significantly reduced memory usage without sacrificing learning speed. It has been confirmed that the memory usage can be reduced by 33% or more when learning using the network ResNet50 used in the field of image recognition. This makes it easier to design larger networks and allows to learn using larger batch sizes in usual networks.

 

2. Chainer’s accompanying array library CuPy has been separated and made into an independent project, allowing a broader range of HPC applications to be easily accelerated using GPUs

The general-purpose array calculation library CuPy is highly compatible with library NumPy, which is very popular in the field of scientific computing, making it possible to run faster using the GPU without altering the code written for use with NumPy. By separating CuPy and developing it as a separate library, we aim to increase users for expanding our application not only in deep learning field but also in other research and development fields.

 

3. Organized the API and made it more intuitive

One of the major features of Chainer is its ability to intuitively describe a complex neural network as a program. We have taken into consideration the various use cases and needs of the community to remove unnecessary options and organize interfaces to provide a more sophisticated API. Due to a more intuitive description, unintentional bugs occur less frequently.

 

● Chainer Release Note: https://github.com/chainer/chainer/releases/tag/v2.0.0

● Chainer Upgrade Guide: https://docs.chainer.org/en/stable/upgrade.html

● Chainer Blog: https://chainer.org/announcement/2017/06/01/released-v2.html

 

The Chainer team plans to release major version updates every four months to support the most advanced research and development activities for researchers and practitioners in each field.

Development results of many external contributors are also included in the Chainer V2 release. PFN will continue to work with support companies and the OSS community to promote the development and dissemination of Chainer.

 

◆ Chainer Meetup # 05

Community event for developers and researchers who use Chainer.

  • Date: June 10, 2017, 14:00–18:30
  • Place: Microsoft Japan Co., Ltd. Shinagawa Office, Seminar Room A
    (Shinagawa Grand Central Tower 31f, 2-16-3 Konan, Minato-ku, Tokyo)
  • Application: https://chainer.connpass.com/event/57307/

 

◆ Kick-off for the Deep Learning Lab Community

The Deep Learning Lab is a community of professionals who are well versed in both technology and business to apply the latest deep learning technology to real business. Microsoft Azure and Chainer are used in key platforms/frameworks to disseminate information about use examples and the latest technology trends.

  • Date: Monday, June 19, 2017, 9:00-12:30
  • Place: Microsoft Japan Co., Ltd. Shinagawa Office
    (Shinagawa Grand Central Tower 31F, 2-16-3 Konan, Minato-ku, Tokyo)
  • Application: https://dllab.connpass.com/event/57981/

 

◆ About the Chainer Open Source Deep Learning Framework (http://chainer.org)

Chainer is a Python-based deep learning framework developed by PFN, which has unique features and powerful performance that enables users to easily and intuitively design complex neural networks, thanks to its “Define-by-Run” approach. Since it was open-sourced in June 2015, as one of the most popular frameworks, Chainer has attracted not only the academic community but also many industrial users who need a flexible framework to harness the power of deep learning in their research and real-world applications.

*Chainer(R) and DIMo(TM) are trademarks of Preferred Networks, Inc. in Japan and other countries.

Drawing app “pixiv Sketch” and automatic coloring service “PaintsChainer” collaborate to provide a new function for automatic coloring of illustrations!

Artificial Intelligence (AI) supports the “coloring” of sketches and illustrations by providing new functions to recognize faces, clothes, and background in the image and automatically filling them with color and shading.

Tokyo, Japan, 24 May 2017 – pixiv Inc. (President: Hiroki Ito, Headquarters: Shibuya-ku, Tokyo) and AI startup Preferred Networks, Inc.  (President & CEO: Toru Nishikawa, Headquarters: Chiyoda-ku, Tokyo, hereinafter referred to as PFN) collaborate to add the new function of automatic coloring, realized by “PaintsChainer”, to the drawing communication platform “pixivSketch”, available from Wednesday, May 24, 2017.

pixiv Sketch is a communication platform that allows users to post drawings directly from devices such as PCs and smartphones. Even when relaxing or playing outside with friends, users can paint anytime and anywhere and experience communication in real-time by posting and sharing their drawings.

The new functionality added to pixiv Sketch is realized using the technology of PaintsChainer that can automatically select painting colors, trained from pairs of line drawings and colored illustrations using Chainer, a deep learning framework developed and provided by PFN.

It allows the user to perform the important process of “coloring” when producing illustrations by selecting a picture drawn on pixiv Sketch or an external image file and then simply clicking the automatic coloring button. Face, clothing, and the background of the illustration are recognized by AI and colors are automatically added. You can also put your favorite color chosen from a color palette as a hint for automatic coloring at any point on the line drawing.

pixiv and PFN will continue to provide valuable services to make drawing and painting more natural and pleasant through AI technology and research.

◆ Automatic coloring function in pixiv Sketch

Release date: May 24th

Cost: free

URL: https://sketch.pixiv.net/ (Available only on Web version)

How to use the new function;

1. Draw a line drawing or select an image of a line drawing

2. Start the automatic coloring tool by pressing the “Automatic coloring” button

3. Select the coloring pattern of your choice from two different styles

4. If necessary, put color hints from the color palette to adjust the coloring

5. After specifying the colors, click the arrow button to complete the coloring process!

 

◆ pixiv Sketch  https://sketch.pixiv.net/

pixiv Sketch is a painting communication platform launched with the desire to “make everyday paintings more casual and fun”. It is a service where you can post images you’ve painted anytime, anywhere through devices such as PCs and smartphones.

 

PaintsChainer   https://paintschainer.preferred.tech/

PaintsChainer is developed and offered by PFN, and received had a great response on Twitter and other social media sites when the service was released in January 2017. Users can upload a black and white drawing file and have it colored automatically using deep learning technology. The user can also supply color hints to control the colorization results.

 

◆ About Preferred Networks, Inc. https://www.preferred-networks.jp/

Founded in March 2014 with the aim of business utilization of deep learning technology focused on IoT. PFN advocates Edge Heavy Computing as a way to handle the enormous amount of data generated by devices in a distributed and collaborative manner at the edge of the network and realizes innovation in the three priority business areas of the transportation system, manufacturing industry, and bio/healthcare.

PFN develops and provides solutions based on the Deep Intelligence in Motion (DIMo) platform that provides state-of-the-art deep learning technology. PFN promotes advanced initiatives by collaborating with world leading organizations, such as Toyota Motor Corporation, Fanuc Inc., National Cancer Research Center.

Preferred Networks and Microsoft have a strategic collaboration in the field of deep learning solutions

Tokyo, Japan, 23 May 2017 – Today, Preferred Networks, Inc. (Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toru Nishikawa, hereinafter referred to as PFN) and Microsoft Corporation (Headquarters: Redmond, Washington, USA, CEO: Satya Nadella) have agreed to strategically collaborate in the field of deep learning solutions with the aim to accelerate the applications of artificial intelligence and deep learning in the business world.

Based on this alliance, both companies will promote cooperation between Microsoft’s public cloud platform Microsoft Azure and PFN’s deep learning technology to provide deep learning solutions for solving problems across a broad range of industries. Microsoft Corporation (Headquarters: Minato-ku, Tokyo, Representative Director: Takuya Hirano) will fully support the delivery of this collaboration to the Japanese market.

 

Through this collaboration, both companies will work together in the following three areas;
1) Technology, 2) Human resource development, 3) Marketing.

1.Technology:

  • Challenges that engineers face in deep learning include the increase of the time required to train complex neural networks, the growing management complexity associated with ever-increasing data, to remain flexible and adaptable to the rapid progress and innovation of deep learning, and the methodology of system development around deep learning. In this collaboration, with the aim of tackling these challenges, both companies will enhance the compatibility between Microsoft Azure IaaS and PFN’s deep learning framework Chainer, providing an Azure template to deploy Chainer/ChainerMN (MN stands for Multi Node) on Azure IaaS with a single click, Chainer to Data Science VMS, Chainer on Azure batch services and SQL Servers, and improving Chainer on Windows by the summer of 2017.

 

  • Currently, the standard way of developing neural networks is to develop from scratch. However, it needs high technical knowledge, and the amount of required investment is also very large. In order to drive the application of deep learning to the real world, it is essential to move from development from scratch to standardized solutions. To realize this transfer, Microsoft Azure Data + Analytics products and PFN’s deep learning platform, Deep Intelligence in-Motion (DIMo) are combined to provide solutions for specific workloads and industries throughout 2017. In addition, both companies will support and nurture partnerships in the development of these solutions to accelerate the broader implementation in the real world.

 

2.Human resource development:

  • The development of data science human resources is one of the main issues of applying deep learning to the real world. In order to resolve this issue, both companies will work together to provide training programs for university students, engineers and researchers throughout 2017. In addition, both companies will consider participation in data science related programs for human resource development, which are typically government organized, for higher education institutions.

 

  • Training programs include not only the basics of neural networks, but also advanced classes that teach how to actually apply deep learning to real-world business applications. Through these training programs, both companies plan to train 50,000 people in three years. As goals for the training, programs such as Imagine Cup and Azure for Research, which are among the world’s largest IT contests for students aiming for fostering international competitive IT talents are considered.

 

3.Marketing:

  • Deep learning is just one method in machine learning, but it is now exposed to many people as a related field of artificial intelligence. As a result, it is difficult for customers to determine whether or not deep learning is effective to solve their business problems. Both companies will start a customer workshop for each industry in the summer of 2017 based on the knowledge of deep learning business cultivated by Microsoft and PFN, and real success stories using Microsoft Azure, Chainer and DIMo.

 

  • By incorporating the latest deep learning technologies provided by Chainer and DIMo on a solid Azure platform, both companies provide an enterprise-grade end-to-end solution that can be incorporated into the customer’s core system, throughout 2017.

 

  • As a place of matching between customers who want to solve business problems with deep learning, and companies who provide consulting services and system development using deep learning, a community named “Deep Learning Lab” has been established, and the community will hold briefings on the days of June 19 and July 25, 2017.
    https://dllab.connpass.com/

Preferred Networks released ChainerMN, a multi-node extension to Chainer, an open source framework for deep learning

Tokyo, Japan, 9 May 2017 –

Today, Preferred Networks, Inc. (Headquarters: Chiyoda-ku, Tokyo, President and CEO: Toru Nishikawa, hereinafter PFN)  released ChainerMN (MN stands for Multi-Node, https://github.com/pfnet/chainermn), which can accelerate the training speed by adding a distributed learning function with multiple GPUs to Chainer, the open source deep learning framework developed by PFN.

Even though the performance of GPUs is continuously improving, the ever-increasing complexity of neural network models, with large number of parameters and much larger training datasets requires more and more computational power to train these models. Today, it is common that one training session takes more than a week on a single node of a state-of-the-art computer.

Aiming to provide researchers with an efficient way to conduct flexible trial and error iterations, while using large training data sets PFN developed ChainerMN, a multi-node extension for high-performance distributed training, built on top of Chainer. We demonstrated that ChainerMN finished training a model in about 4.4 hours with 32 nodes and 128 GPUs which would require about 20 days on a single-node, single GPU machine.

 

  • Performance comparison experiment between ChainerMN and other frameworks

https://research.preferred.jp/2017/02/chainermn-benchmark-results/

We compared the performance benchmark result of ChainerMN with those of other popular multi-node frameworks. In our 128-node experiments with a practical setting, in which the accuracy is not sacrificed too much for speed, ChainerMN outperformed other frameworks.

 

When comparing the scalability, although the single-GPU throughputs of MXNet and CNTK (both are written in C++) are higher than ChainerMN (written in Python), we found that the throughput of ChainerMN was the highest with 128 GPUs, showing that ChainerMN is the most scalable. This result was due to the design of ChainerMN which is optimized for both intra-node and inter-node communications.

 

Existing Chainer users can easily benefit from the performance and scalability of ChainerMN simply by changing a few lines of their original training code.

ChainerMN has already been used in multiple projects in a variety of fields such as natural language processing and reinforcement learning.

 

  • About the open source deep learning framework Chainer

Chainer is a Python-based deep learning framework developed by PFN, which has unique features and powerful performance that enables users to easily and intuitively design complex neural networks thanks to its “Define-by-Run” feature. Since it was open-sourced in June 2015, as one of the most popular frameworks, Chainer has attracted not only the academic community but also many industrial users who need a flexible framework to harness the power of deep learning in their research and real-world applications. (http://chainer.org/

 

  • About Preferred Networks, Inc.

Founded in March 2014 with the aim of business utilization of deep learning technology focused on IoT. Edge Heavy Computing handles the enormous amount of data generated by devices in a distributed and collaborative manner at the edge of the network and realizes innovation in the three priority business areas of the transportation system, manufacturing industry, and bio- healthcare.

PFN develops and provides solutions based on the Deep Intelligence in-Motion (DIMo, Daimo) platform that provides state-of-the-art deep learning technology. Collaborating with world leading organizations, such as Toyota Motor Corporation, Fanuc Inc., National Cancer Research Center,  we are promoting advanced initiatives.(https://www.preferred-networks.jp/en/

 

*Chainer(R) and DIMo(TM) are a trademark of Preferred Networks, Inc. in Japan and other countries.

A member of Preferred Networks temporally assigned to OpenAI

On April 10th 2017, Yasuhiro Fujita, an engineer of Preferred Networks, starts working at OpenAI as a visiting research scientist in San Francisco, California.

OpenAI is a non-profit artificial intelligence research company, in which many famous researchers are working on the fundamental problems in AI, and making the intellectual property and research open to public.

Yasuhiro Fujita is a passionate researcher of game-playing AI, and the creator of ChainerRL, a deep reinforcement learning library on top of Chainer.

During his assignment until September, Fujita will be contributing to OpenAI’s research projects and later making his achievements public.

Preferred Networks, as an active member of the research community, will keep its strong commitment to the advancement of the research in this field throughout similar collaborations and initiatives.

Intel and Preferred Networks collaborate to jointly develop Chainer, deep learning open source framework

The companies aim to significantly accelerate CPU performance for Chainer running on Intel Architecture.

Intel Corporation and Preferred Networks Inc. (PFN) announced today that the companies will collaborate on the development of Chainer(R)(http://chainer.org/), PFN’s open source framework for deep learning, with the aim to accelerate out of the box deep learning performance on general purpose infrastructure powered by Intel.

 

Advanced technologies including IoT (Internet of Things), 5G (fifth generation mobile networks), and AI (artificial intelligence) are expected to be used in a range of industries ahead, giving rise to data-driven business opportunities and user experiences. The advance of technologies related to AI and deep learning, in particular, will accelerate the creation of applications that further enhance the intrinsic value of data.

The use of special-purpose computing environments for developing and implementing AI applications and deep learning frameworks poses challenges for the developer community, including development complexity, time and cost.

PFN, the developer of Chainer, which is an advanced deep learning framework with a reputation for ease of use among application developers in various industries, and Intel Corporation, a provider of general purpose computing technologies and industry leading AI/deep learning accelerators, will collaborate in an effort to make AI development easier and more affordable. The collaboration will bring both companies’ technologies to bear in the aim of optimizing development/execution of applications that use advanced AI and deep learning frameworks, as well as accelerating the performance of image and voice recognitions.

Chainer is a Python-based deep learning framework developed by PFN, which has unique features and powerful performance that enables users to easily and intuitively design complex neural networks thanks to its “Define-by-Run” feature. Since it was open-sourced in June 2015, as one of the most popular frameworks, Chainer has attracted not only the academic community but also many industrial users who need a flexible framework to harness the power of deep learning in their research and real-world applications.

Intel Corporation, a technology leader that is uniquely poised to drive the AI computing era, will help Chainer deliver breakthrough deep learning throughput for the industry’s most comprehensive compute portfolio for AI, which includes the Intel(R) Xeon(R) processors, Intel(R) Xeon Phi™ processors, Intel(R) Arria(R) 10 FPGAs, Intel(R) Nervana™ technology and more products. This framework will employ the highly-optimized Intel’s open source library—Intel(R) Math Kernel Library (MKL) and Intel(R) Math Kernel Library Deep Neural Network (MKL-DNN) as a fundamental building block.

 

Through the collaboration, Intel and PFN will undertake the following.

  •  Continuously optimize the performance of Chainer on Intel architecture
  •  Continuously align to Chainer updates
  •  Continuously optimize Chainer to updates to Intel architectures for general purpose computing, accelerators, libraries, and so on
  •  Share the results of the companies’ collaboration with the community on Intel’s GitHub repository
  •  Collaborate on marketing activities designed to accelerate AI/deep learning market growth

 

*Intel, Xeon, Xeon Phi, Arria and Nervana are trademarks or registered trademarks of Intel Corporation in the United States and other countries.

*Chainer, DIMo are trademarks or registered trademarks of Preferred Networks, Inc. in Japna and other countries.

PFN 2017 Summer Internship Program

As goes the tradition, Preferred Networks (PFN) will be organizing the internship program this summer too. From this year, we are also looking for front-end/back-end and chip development in addition to machine learning. We welcome applications not only from machine learning field but also from many people. We are looking forward to receiving students who want to join us in creating new technologies and services. Students who previously applied are also welcome to try again this year.
(Application from overseas with a need for VISA is already closed for this year)

Application Guideline

 

●Period

 

August 1st – September 30th 2017
(Negotiable.)

 

●Time & Place

8 hours/day, 5 days/week (excluding holidays)
Otemachi-Bldg. 2F 1-6-1, Otemachi, Chiyoda-ku,Tokyo, 100-1004

 

●Salary

 

  • High school: 1500Yen/hour
  • Technical college/Undergraduate/Graduate: 1800Yen/hour
  • Transportation expenses (up to 10000Yen/month) are also covered.

 

●Why join the PFN internship program?

 

  • You will be collaborating and be mentored by experts in various fields including deep learning, computer vision, natural language processing, reinforcement learning, algorithms, distributed processing, etc.
  • You can make public the results of your work during the internship program, as OSS or a paper, etc. (Some restrictions might apply.)

 

●Qualification requirements

We are looking for highly motivated people who have development capabilities. Expertise in the fields mentioned below, or prior development experience are taken into consideration, but are not a must. Application requirements are as follows:

  • Currently students (High school, technical college, college, graduate students, others could also be discussed.)
  • Able to communicate in English or Japanese
  • Able to communicate on one’s own initiative
  • Have programming skills (regardless of the programming language)
  • Able to work fulltime on weekdays at our Tokyo office

# We will prepare accomodation for those who live far from Tokyo.
# You can still apply even if you are not a fully-fledged application developer.

 

●How to apply

Please submit the application form below.
https://docs.google.com/forms/d/e/1FAIpQLSevjHAtBhq9380kzDLXQ1dySoWa_p7N_VhgTHZnC4pcJa75hw/viewform

Questions about the internship program are also accepted by intern2017@preferred.jp.

Application form note

Proof of skills; upload your document following the steps below that explains your strengths and expertise fields, etc. (Microsoft Word or Google docs, one A4 page)
E.g., List of papers, received awards, developed/used Software&Services, programming contests participation history, personal website/blog, twitter account, etc.
https://www.preferred-networks.jp/wp-content/uploads/2017/03/intern2017_GoogleUpload_3.pdf

Themes you want to do; please include your interest in the selected themes and your expectations from the internship using less than 400 characters.

# This is a very important for both the admission process, and the internship theme selection.

 

●Application Deadline

 

May. 7th, 2017 23:59 (JST)

 

●Selection process

Documents screening
# Takes around one weeks before result is returned.

Pre-interview task screening
# The task will be announced to those who passed the above.

Interview (generally once)
# Skype interview for remote applicants

Acceptance notice (Late June)

 

●Themes

 

[Machine Learning / Mathematics Fields]

Applications

  • Chainer development
  • Image recognition
  • Video analysis
  • Content generation (Generation of images, videos, sounds, etc.)
  • Natural language processing
  • Speech recognition
  • Anomaly detection
  • IoT
  • Data compression
  • Robotics (Robot arms, bipedal walking, self-driving cars, path planning)
  • Genomics, Epigenomics, proteomics
  • Deep Learning on embedded systems
  • LSI design optimization

 

Research

  • Distributed algorithm, Distributed deep learning
  • Reinforcement learning
  • Optimization
  • Deep generative models
  • Model compression
  • Neural network quantization
  • Machine learning with limited labels (One-shot learning, Weakly supervised learning, Semi-supervised learning, Meta learning)
  • Machine learning using simulators
  • Interpretability in machine learning
  • Differential privacy
  • Communication or collaboration emergence

 

[Front-end or Back-end Development]

  • Chainer development
  • SensorBee
  • PaintsChainer
  • Stream processing
  • Tools development
  • Web development
  • Networking
  • High-performance computing
  • 3DCG
  • Unity development
  • AR or VR

 

[Chip Development]

  • FPGA design

Preferred Networks has been awarded with Technology Award in the Financial Times ArcelorMittal Boldness in Business Awards 2017

Preferred Networks has been received with the Technology Award in the FT ArcelorMittal Boldness in Business Awards 2017, held in London, on March 17th.

IoT and deep learning technology developed by PFN has been highly evaluated in the fields of collaborations with FANUC for factory automation and diagnosis of industrial robots, and Toyota Motors for self-driving cars.

    ●the FT ArcelorMittal Boldness in Business Awards;
    This awards was established in 2008 by ArcelorMittal, world’s biggest steel and mining company and Financial Times. This year’s Boldness in Business had six category awards, and nominate the innovative companies around the world.
    Winners are selected by a panel of experts including Lionel Barber, Editor of the Financial Times and Lakshmi Mittal, Chairman & Chief Executive of ArcelorMittal, who base their decision on a shortlist drawn up by the FT’s global network of bureau chiefs and senior journalists.

●Detail of the Award and 2016 winners
https://live.ft.com/Events/2017/FT-ArcelorMittal-Boldness-in-Business-Awards

●the ceremony in London
https://www.flickr.com/photos/45442848@N05/33110753460/in/album-72157678023721663/

●Preferred Networks teaches robots to collaborate through deep learning
https://www.ft.com/content/a1f3c3a0-d368-11e6-b06b-680c49b4b4c0

SORACOM and Preferred Networks carry out Co-demonstration of “Edge Heavy Computing” using machine learning technology on IoT equipment at “CeBIT 2017” in Hannover Germany

Preferred Networks, Inc. (Headquarters: Chiyoda-ku, Tokyo; President and CEO: Toru Nishikawa, hereinafter PFN), and SORACOM,INC(Headquarters: Setagaya Ward, Tokyo; President and CEO: Ken Tamagawa), announce that at the CeBIT 2017 International Information and Communication Technology Trade Fair, to be held in Hanover Germany from 20th to 24th March, we will carry out a joint demonstration using the deep learning technology developed by PFN on IoT equipment connected to Soracom’s network to showcase the concept of “Edge-Heavy Computing.”

The “SORACOM” IoT communications platform delivers secure, scalable over-the-air connectivity purpose-built for IoT, and supports rapid deployment and operation of IoT systems. With the cloud-native “SORACOM” platform, IoT connectivity can be effectively secured from end to end, integrated natively with leading cloud services, and managed directly via Web console or API.

PFN has strengths in machine learning technologies, including IoT and deep learning. PFN develops and provides software platform products that enable data analytics in distributed and collaborative manner to realize advanced intelligence in the fields of manufacturing, transportation, and bio-healthcare.

With wireless technology such as cellular line provided by Soracom, it is becoming easier to send data directly from the IoT devices to the cloud. However, considering bandwidth and data volume, there is still a challenge to continuously transmit large-scale data such as video and audio in real time.

The demonstration of this time is to make it possible to solve these problems by executing deep learning technology and analyze the data directly on the IoT device, transmitting only the truly valuable data to the cloud. PFN calls this technology “edge heavy computing.” By delivering real-time analysis at the edge, it is possible to solve challenges related to privacy and bandwidth, minimizing data volume required for transmission and discarding video data immediately after analysis.

Co-demo of “Edge Heavy Computing” Details:

1. Content of the demo
We will demonstrate PFN’s deep learning platform “DIMo (Deep Intelligence in-Motion)” operating on “NVIDIA® Jetson™ TX 1”, a high-performance embedded GPU module. Then, we will analyze the demographics of individuals observed by a camera connected to “NVIDIA Jetson TX 1” on the spot. After that, we will forward only the summarized information to the cloud using “SORACOM Air,” and visualize it using “SORACOM Harvest”.

2. Benefits of “Edge Heavy Computing”

The merits obtained by transferring only rough information such as age and gender of the person captured by the camera and position information in the image to the cloud are as follows:

      ●When sending images to the cloud directly, resolution and quality have to be dropped, increasing the difficulty of analysis

 

      ●Compared to the typical case of sending images directly to the cloud for analysis, it is only necessary to send very small amounts of data

 

      ●Analysis of high quality and high resolution video without can be conducted without concern for data size

 

      ●Since images are not accumulated in the cloud, they are easily discarded on the camera side after analysis and privacy can be easily maintained

 

3. Exhibit of joint demonstration Details

・Time: 20th – 24th March 2017
・Venue: Hannover, Germany “CeBIT 2017 (International Information and Communication Technology Trade Show)”
・Booth number: Hall 12, Stand B 37 http://www.cebit.de/exhibitor/soracom-dk/U177198

* Soracom will also exhibit in Japan Pavillion, Hall 4, Stand A 38, but the above demo exhibition will be Hall 12 only.

 

NVIDIA offered edge devices for this demonstration exhibition of “Edge Heavy Computing”

Mr. Masataka Osaki, NV representative, Japan representative and vice president of US head office

“As an AI computing company, NVIDIA offers end-to-end solutions from learning server side to deep learning to edge side inference.” The demonstration of this “edge heavy computing” will embody the deep learning solution and will continue to support the efforts of both companies.”

demo1

demo2

About SORACOM,INC
Soracom Corporation provides a communication platform, “SORACOM,” for IoT. This platform, launched in September 2015, is a mobile communication service specialized for IoT. By using SORACOM, customers can incorporate communication essential to the IoT system reasonably, securely and programmably into the system.

About Preferred Networks, Inc.
Founded in March 2014 with the aim of business utilization of deep learning technology focused on IoT. Edge Heavy Computing handles the enormous amount of data generated by devices in a distributed and collaborative manner at the edge of the network and realizes innovation in the three priority business areas of the transportation system, manufacturing industry, and bio- healthcare.

PFN develops and provides solutions based on the Deep Intelligence in-Motion (DIMo, Daimo) platform that provides state-of-the-art deep learning technology. Collaborating with world leading organizations, such as Toyota Motor Corporation, Fanuc Inc., National Cancer Research Center, we are promoting advanced initiatives.

About NVIDIA Jetson TX 1
Jetson TX 1 is a supercomputer installed in a credit card size module. NVIDIA Maxwell ™ architecture, 256 NVIDIA CUDA® core, 64-bit CPU with very efficient processing power. It also has the latest technology of deep learning, computer vision, GPU computing and graphics, making it the most suitable module for embedded visual computing.

NVIDIA Jetson TX 1

PFN members gave a tutorial on deep learning implementations at AAAI-17

[San Francisco, February 5th] Preferred Networks members, Seiya Tokui and Kenta Oono, gave a tutorial titled “”Deep Learning Implementations and Frameworks (DLIF)” at an international conference (AAAI-17).

Based on the fact that using software frameworks is fundamental in deep learning applications, the purpose of this tutorial is to help users to select an appropriate deep learning framework by describing the basics of implementation, design choices, and comparison of the features of the existing frameworks including Chainer,

The presentation slides and sample code can be found here.

AAAI has more than 30 years history as a prestigious academic conference in artificial intelligence. It hosted 24 tutorials this year, with diverse topics from machine learning theory to AI applications to IoT or robotics. The DLIF tutorial attracted the largest number of pre-registrants out of them. This work was co-organized by Dr. Atsunori Kanemura of AIST (National Institute of Advanced Industrial Science and Technology in Japan), also under supervision from Dr. Toshihiro Kamishima and Dr. Hideki Asoh.




(From right to left, Seiya Tokui of PFN, Kenta Oono of PFN and Dr.Atsunori Kanemura of AIST )

Preferred Networks will continue contributing to academia through open source software, research papers, and tutorial talks.

[Closed] 2nd Call for application: 2017 summer internship in Tokyo

Preferred Networks will be organizing internship programs next summer in Tokyo. In order to make the process smooth for the students outside of Japan, we open an early bird application opportunity for the first time. We are looking forward to welcoming students who want to join us in creating new technologies and services. Note that similar programs will follow for the students both in & outside of Japan, also after this first call.

Target of this program

– Students outside of Japan

Work time & Location:

Business hours:
8 hours/day, 5 days/week (excluding national holidays)
Location: Center of Tokyo
Preferred Networks Tokyo office: Otemachi Bldg. 2F, 1-6-1, Otemachi, Chiyoda-ku, Tokyo, Japan 100-0004
https://www.preferred-networks.jp/en/about


Period & Compensation:

– The period of the internship can be flexibly arranged though it is usually held during July & August
– We require minimum of two month (40 business days), in order to be able to tackle a challenging task
– Interns are paid a competitive salary
– Residence and travel cost is to be provided


Requirements:

– Formally enrolled in university or research institute outside of Japan during 2017-2018 school year
– Fluent in English (or Japanese)
– Good programming skill (any programming language)
– Computer science basics
– Able to work full-time on weekdays at our Tokyo office during the period


Preferred experience & skills:

– Machine learning and deep learning basics
– Experience with software & service development
– Experience with team development
– Contribution to open source projects


Candide themes (subject to change)

1. Technology areas: Sub-field of machine learning, such as
Deep learning theory
Reinforcement learning
Computer vision
Parallel distributed learning
Weakly supervised learning
Transfer learning
Anomaly detection
Deep Generative Model
Others
2. Application areas: Advanced IoT applications, such as
Image recognition
Robotics & machine control
Life science & medicine
Machine Learning Framework development (Incl. OSS such as Chainer)
Others

(FYI) Projects of 2016 summer interns
DQN with Differentiable Memory Architectures
Multi-modal Deep Generative Model for Anomaly Detection
CNN based robotic grasping for randomly placed objects by human demonstration
Anomaly Detection by ADGM / LVAE
Imitation Learning for Autonomous Driving in TORCS
3D Volumetric Data Generation with Generative Adversarial Networks
Bayesian Dark Knowledge and Matrix Factorization
Automatically Fusing Functions on CuPy
Generation of 3D-avatar animation from latent representations
Response Summarizer: An Automatic Summarization System of Call Center Conversation
Product marketing in conversations


Application information:

– Resume / CV (PDF format only. Please DO NOT include any private information e.g. age, personal address, phone number, etc.)
– Name, e-mail address, affiliation
– Github account (optional)
– Linked.in account (optional)


How to apply:

– Please fill the google form. (Application is now closed and no e-mail application will be accepted)
Due: January 20th, 11:59pm Friday (PST)
– No late submission will be accepted
– The interview process takes about 2-4 weeks after application submission
– Usually, getting visa support in Japan takes up to 3 months so that the preparation must be done in advance of the internship period


Interview process:

1. Document review
2. Skype interviews in English or Japanese (multiple times if necessary)

If you have questions, please contact us at hr-pfn@preferred.jp (Sorry but no late application is accepted for fairness)

[Closed] Call for application: 2017 summer internship in Tokyo, Japan

Note: this application is now closed. We are planning to have one more turn early January.

Preferred Networks will be organizing internship programs next summer in Tokyo. In order to make the process smooth for the students outside of Japan, we open an early bird application opportunity for the first time. We are looking forward to welcoming students who want to join us in creating new technologies and services. Note that similar programs will follow for the students both in & outside of Japan, also after this first call.

Target of this program

– Students outside of Japan

Work time & Location:

Business hours:
8 hours/day, 5 days/week (excluding national holidays)
Location: Center of Tokyo
Preferred Networks Tokyo office: Otemachi Bldg. 2F, 1-6-1, Otemachi, Chiyoda-ku, Tokyo, Japan 100-0004
https://www.preferred-networks.jp/en/about


Period & Compensation:

– The period of the internship can be flexibly arranged though it is usually held during July & August
– We require minimum of two month (40 business days), in order to be able to tackle a challenging task,
– Interns are paid a competitive salary
– Residence is to be provided


Requirements:

– Formally enrolled in university or research institute outside of Japan during 2017-2018 school year
– Fluent in English (or Japanese)
– Good programming skill (any programming language)
– Computer science basics
– Able to work full-time on weekdays at our Tokyo office during the period


Preferred experience & skills:

– Machine learning and deep learning basics
– Experience with software & service development
– Experience with team development
– Contribution to open source projects


Candide themes (subject to change)

1. Technology areas: Sub-field of machine learning, such as
Deep learning theory
Reinforcement learning
Computer vision
Parallel distributed learning
Weakly supervised learning
Transfer learning
Anomaly detection
Deep Generative Model
Others
2. Application areas: Advanced IoT applications, such as
Image recognition
Robotics & machine control
Life science & medicine
Machine Learning Framework development (Incl. OSS such as Chainer)
Others

(FYI) Projects of 2016 summer interns
DQN with Differentiable Memory Architectures
Multi-modal Deep Generative Model for Anomaly Detection
CNN based robotic grasping for randomly placed objects by human demonstration
Anomaly Detection by ADGM / LVAE
Imitation Learning for Autonomous Driving in TORCS
3D Volumetric Data Generation with Generative Adversarial Networks
Bayesian Dark Knowledge and Matrix Factorization
Automatically Fusing Functions on CuPy
Generation of 3D-avatar animation from latent representations
Response Summarizer: An Automatic Summarization System of Call Center Conversation
Product marketing in conversations


Application documents:

– Resume (CV)


How to apply:

– Application has been closed
Due: November 23rd, 11:59pm (PST)
– The interview process takes about 2-4 weeks after application submission
– Usually, getting visa support in Japan takes up to 3 months so that the preparation must be done in advance of the internship period


Interview process:

1. Document review
2. Skype interviews in English or Japanese (multiple times if necessary)

Use case of Chainer Playground in Broadband Tower Space Weather Forecast research group

The BroadBand Tower Space Weather Forecast research group at Kyoto University is performing research on predicting solar activities that affect the human civilization every day, by applying Deep Learning technologies to a large amount of astronomical measurement data, collected in a cooperative effort between Kyoto University and people all over the world. (https://www.usss.kyoto-u.ac.jp/bbt.html)

The group has conducted a series of lectures on the prediction of space weather based on Chainer ((http://chainer.org), a Deep Learning framework developed by Preferred Networks. During the lectures, using an early access version of Chainer Playground, a Chainer environment that can be used via a web browser, the students can learn the basics of Deep Learning and the usage of Chainer without installing Python and the various numerical libraries used by Chainer on their laptops.

Takayuki Muranushi, adjunct lecturer at Kyoto University, commented: “This online self-studying environment is very helpful, since it does not require the environmental setup that usually takes a few weeks, and students can use it regardless of the operating system and performance of their laptops.”

PFN recognized as one of most innovative and exciting Japanese startups at 2016 US-Japan Innovation Awards

Toru Nishikawa, CEO, receives “Innovation Showcase” award at the symposium

Preferred Networks recognized as one of most innovative and exciting Japanese startups at 2016 US-Japan Innovation Awards

Palo Alto, CA — July 22nd, 2016 — Preferred Networks, Inc. (PFN) was selected one of the five “Innovation Showcase” companies at 2016 US-Japan Innovation Awards Symposium held in the Stanford University Arrillaga Alumni Center.

US-Japan Innovation Awards is operated by the Japan Society of Northern California in collaboration with the Stanford University US–Asia Technology Management Center. The “Innovation Showcase” is given to most exciting Japanese startup companies that have potential to disruput or transformational globally and to be known to the Sillicon Valley.

 

About Preferred Networks

Preferred Networks Inc. (PFN) is a Tokyo-based startup focusing on applications of latest artificial intelligence technologies to emerging problems in the Internet of Things (IoT). PFN’s vision is the realization of Deep Intelligence – a future IoT in which all devices, as well as the network itself, are intelligent. PFN develops software related to deep learning and IoT, and provides Deep Intelligence in-Motion solutions. PFN collaborates with many world-leading companies in industries, such as FANUC for intelligent robots and Toyota motors for autonomous driving. For more information please visit: www.preferred-networks.jp/en

PFN achieves Second Place at “Pick Task” of Amazon Picking Challenge

 
 
A team of Preferred Networks researchers and engineers participated in the Amazon Picking Challenge from June 29 to July 3, 2016, at Leipzig, Germany, and achieved 2nd place (score tie with 1st place) in the “pick task” and 4th place in the “stow task”.

The Amazon Picking Challenge is a competition with the objective to build a robot that can automate typical warehouse tasks without human intervention. In the “pick task” the objective is to take 12 specified items out of a shelf that also contains around 46 other items and put them into a box. In the “stow task” the objective is to take all 12 items from a box and put them into a shelf that already contains around 34 other items. In these tasks, the robot must operate without any human control. The Challenge combines object recognition, pose recognition, grasp planning, and motion planning.

The PFN Team used state-of-the-art Deep Learning algorithms and Chainer, a Python-based Open Source Deep Learning framework, on input data obtained from image and 3D location sensors for object detection and localization of the best approach position for each object. On the hardware side, the team used two FANUC robot arms of type M-10iA, equipped with self-built specialized end effectors to grab items of various characteristics reliably, then built the robot motion control from scratch.

Expertise in these technologies enabled Preferred Networks to build a robot that can compete with the world’s leading robotics research groups and companies in only three months’ time and is an important milestone on the way to revolutionize the industrial IoT and robotics.

References:

PFN will participate in Amazon Picking Challenge 2016

Preferred Networks will participate in the Amazon Picking Challenge 2016 (http://amazonpickingchallenge.org/) from June 29 to July 3 in Leipzig, Germany.

The Amazon Picking Challenge is a competition with the objective to build a robot that can recognize and take out items of various shapes and materials from a shelf and put them into a box – and the other way around – without human intervention. Our team will use Deep Learning technology for visual object recognition in conjunction with multiple robot arms that have specialized sensors and end effectors.

Preferred Networks is conducting research and development on Deep Learning to revolutionize the industrial IoT, and participating in this challenge is one step on the way to achieve this goal.

Manufacturing Automation Leaders Collaborate: Optimizing Industrial Production through Analytics

TOKYO, — April 18, 2016 —FANUC CORPORATION, the world’s leading supplier of robotics and factory automation, is collaborating with Cisco, the worldwide leader in IT-enabled digitization, as well as Rockwell Automation, the world’s largest company dedicated to industrial automation and information solutions, and Preferred Networks, a leading provider of Artificial Intelligence solutions. The companies will work together on the development and deployment of the FANUC Intelligent Edge Link and Drive (FIELD) system, a platform that connects not only Computer Numerical Control machines (CNCs) and robots, but also peripheral devices and sensors – to deliver analytics that optimize manufacturing production.

The FIELD system will be a platform for the delivery of advanced analytics for FANUC CNCs, robots, peripheral devices, and sensors used in automation systems. It will drive improved machine reliability, quality, flexibility, and speed – elevating the Overall Equipment Efficiency (OEE) and increasing manufacturing profitability. It will also provide advanced machine learning and deep learning capabilities. By working with Cisco, Rockwell Automation, and Preferred Networks – FANUC will offer a complete solution including network and compute infrastructure, applications, and an enabling middleware platform. With this open platform, application developers, sensor and peripheral device makers, system integrators and others can build solutions that improve equipment efficiency, manufacturing output, and quality.

Extension of Current Successes
The FIELD system extends the success of the existing FANUC ZDT (Zero Downtime) connected robots project; ZDT is built with Cisco cloud, IOT data collection software, and end to end security capabilities. The companies are working together to implement systems for major automotive manufacturers by leveraging the Allen-Bradley Stratix Ethernet switches from Rockwell Automation to connect robots to a Cisco Unified Computing System (UCS) – all of this running on FANUC and Cisco’s ZDT Data Collection software. Automotive customers who have implemented this system are quickly realizing a significant decrease in production downtime as well as increased cost savings.

Intelligent Manufacturing
The FIELD system will bring the power of advanced machine learning and artificial intelligence into the hands of customers and application developers — driving manufacturing to a new level of productivity and efficiency. FANUC and Preferred Networks have established these new technologies for applications such as Bin Picking robots, anomaly detection, and failure prediction. The FIELD system combines both artificial intelligence and edge-computing technologies – enabling distributed learning. Data is generated from robots and machines and is processed in real time at the edge of the network. This allows those devices to intelligently coordinate and collaborate in a flexible manner – resulting in sophisticated manufacturing practices not before possible.

 

Who provides What for the FIELD System?

FANUC: Provides CNCs and robots with embedded sensors to track key variables required to improve machine reliability, quality, and speed.

FANUC, Cisco and Preferred Networks: Provides enabling middleware Platform Software, as well as security and application lifecycle management applications. It will also leverage Preferred Networks’ open deep learning framework (Chainer), IoT Stream Engine (SensorBee) – and other advanced machine learning libraries within its Deep Intelligence in-Motion (DIMo) platform.

Cisco and Rockwell Automation: Provides networking, compute and security infrastructure to connect the robots, CNCs, and other cell equipment to the FIELD applications. Based on the Rockwell Automation and Cisco collaboratively developed Converged Plantwide Ethernet (CPwE) architecture – this further drives improved security, connectivity, flexibility, and scalability – all of which allows connection from a single, small cell to large factory with hundreds of cells.

FANUC, Rockwell Automation and Preferred Networks: Delivers the initial Application Software on top of the FIELD middleware and infrastructure platform. This will extend the LINK-i and ZDT applications currently deployed, along with additional deep learning applications from the FANUC and Preferred Networks partnership. It is planned that Rockwell Automation manufacturing software products (including FactoryTalk View, FactoryTalk VantagePoint, and FactoryTalk Production Center) will seamlessly integrate with the FIELD system to speed deployment.

 

LEADERSHIP QUOTES

Sujeet Chand, Senior VP and CTO, Rockwell Automation
“By collaborating with world class companies, Rockwell Automation helps maximize manufacturers’ investments by leveraging the data from their intelligent devices they are using today to drive an enterprise-wide analytics strategy. With a secure scalable compute approach to analyze this data – from device to the enterprise – users can improve operations and make more informed decisions tailored to meet the needs of their organizations.”

Rowan Trollope, SVP IoT and Applications, Cisco
This collaboration represents a historic shift in the industry, with IoT, industrial automation and machine learning coming together to make the factory of the future a reality. It’s been talked about for years, but now it is really happening. Cisco couldn’t be more thrilled to be a part of this effort, one that will be key to our positioning in other industries that want to realize the benefits of digitization.”

Toru Nishikawa, CEO, Preferred Networks
“PFN is excited that this collaboration will further accelerate the advancement of the manufacturing industry. Since the start of our work with FANUC, leveraging machine learning and artificial intelligence has been aimed not only at making machines and robots smarter, but also towards a continuous improvement of manufacturing productivity through intelligent real-time coordination and collaboration between robots and machines. We are confident that FIELD will play a central role in making that vision a reality.

  

About FANUC CORPORATION
FANUC CORPORATION, headquartered at the foot of Mt. Fuji, Japan, is the global leader and the most innovative manufacturer of Factory Automation, Robots and ROBOMACHINE’s in the world. With 252 offices in 46 countries, FANUC provides world-class service and support to customers globally. Since its inception in 1972, FANUC has contributed to the automation of machine tools as a pioneer in the development of computer numerical control equipment. FANUC technology has been a leading force in a worldwide manufacturing revolution, which evolved from the automation of a single machine to the automation of entire production lines. For more information visit: http://www.fanuc.co.jp/eindex.htm.

About Cisco
Cisco (NASDAQ:CSCO) is the worldwide leader in IT that helps companies seize the opportunities of tomorrow by proving that amazing things can happen when you connect the previously unconnected.
For ongoing news, please go to http://thenetwork.cisco.com.

About Preferred Networks
Preferred Networks Inc. (PFN) is a Tokyo-based startup focusing on applications of latest artificial intelligence technologies to emerging problems in the Internet of Things (IoT). PFN’s vision is the realization of Deep Intelligence – a future IoT in which all devices, as well as the network itself, are intelligent. PFN develops software related to deep learning and IoT, and provides Deep Intelligence in-Motion solutions. PFN collaborates with many world-leading companies in industries, such as FANUC for intelligent robots and Toyota motors for autonomous driving. For more information please visit: www.preferred-networks.jp/en

PFN & PFI Summer Internship Program

As goes the tradition, Preferred Infrastructure (PFI) and Preferred Networks (PFN) will be organizing the joint internship program this summer too. We are looking forward to receiving students who want to join us in creating new technologies and services. Students who previously applied are also welcome to try again this year.

Application Guideline
Qualification requirements and emplyment conditions are same as PFN and PFI, except that you must be as fluent in Japanese as natives if you want to work on PFI’s themes. To which company candidates will be assigned depends on the themes they will work on.

Period
August 1st 2016 – September 30th 2016
(The period of the internship can be flexibly arranged.)

Time & Place
8 hours/day, 5 days/week (excluding holidays)
Otemachi-Bldg. 2F 1-6-1, Otemachi, Chiyoda-ku,Tokyo, 100-1004

Salary
High school: 1200Yen/hour
Technical college/Undergraduate/Graduate: 1500Yen/hour
Transportation expenses (up to 10000Yen/month) are also covered.

Why join the PFN internship program?
・You will be collabolating and be mentored by experts in various fields including information retrieval, natural language processing, deep learning, algorithms, distributed processing, computer vision, etc.
・You will experience every aspect of a startup company. It is valuable to those who want to work in or to be a founder of startups.
・You can make public the results of your work during the internship program, as OSS or paper, etc. (Some restrictions might apply)

Qualification requirements:
We are looking for highly motivated people who have Software development capabilities. Expertise in the fields mentioned above, or prior development experience are taken into consideration, but are not a must. Application requirements are as follows:
・Currently a students 18 of age or older (High school, technical college, college, graduate students, others could also be discussed.)
・Have programming skills (regardless of the programming language)
・Able to work fulltime on weekdays at our Tokyo office
・If you want to work on PFI’s themes, you must be as fluent in Japanese as natives (fluency in English and Chinese are welcome).
# We will prepare accommodation for those who live far from Tokyo.
# Women and Non-Japanese speakers are also very welcome
# You can still apply even if you are not a fully-fledged application developer

How to apply:
Please send the application documents below to intern-apply@preferred.jp
Questions about the internship program are also accepted by this address.

Application documents:
・Resume (Name, address, contact information, background)
・Proof of skills; explain your strengths and expertise fields, etc. (one A4 page)
Ex: List of papers, received awards, developed/used Software&Services, programming contests participation history, personal website/blog, twitter account, etc.
・Motivation letter (Please include your interest field/theme and your expectations from the internship.)
# This is a very important for both the admission procces, and the internship theme selection, so please describe in as much detail as possible.

Application Deadline
May. 8th, 2016 (sun.) 23:59 (JST)

Selection process:
Selection process:
1. Application documents screening: Starts on May 8th. Results are announced to applicants via email in a week.
2. Pre-interview task: Applicants who passed receive a task via email.
3. Interviews: First interview starts in June (Skype is used in case of remote applicants)
4. Final results: Result of the selection processes is announced to applicants in the beginning of July.

Themes
@PFI
1.Technology field: Natural Language Processing for Japanse documents
・Automatic document classification
・Named entity recognition
・Synonym Extraction
・Automatic summarization
・Similarity search

2. Technology field: Speech Recognition for Japanese speech
・Speech recognition by deep learning.
・Document recognition by deep learning.
・Speaker identification
・Noise filtering

@PFN

Application fields
・Image recognition
・Anomaly detection
・Robotics (bipedal walking, car control)
・Genomics, Epi-genomics, proteomoics
・Art generation (Generation of images, videos, sounds etc.)
・Application of deep learning to embedded system
Stream processing for IoT

Research fields
・Machine learning with limited labels (One-shot, Weakly superivsed, Semi-supervised Learning)
・Distributed algorithm, Distributed deep learning
・Deep generative model
・Machine learning using simulators

Change of Address

This is to inform you that we will be moved to the below address as of April 25, 2016.

[New location]
Otemachi-Bldg.2F
1-6-1, Otemachi
Chiyoda-ku,Tokyo, 100-1004

From April 25th, we will do normal operation.
Thank you for your understanding and support.

FEB18 Shohei Hido Speaks at JAPAN-UK Robotics and Artificial Intelligence Seminar @ London

Shohei Hido, Chief Research Officer of Preferred Networks, participated in a Japan-UK Robotics and Artificial Intelligence Seminar 2016 that was held on February 18th in London. The event was co-organized by the British Embassy in Tokyo and the Japan Embassy in London in order to discuss and enhance the collaboration between the research communites and govermental agencies in Japan and the United Kingdom.

From Japan, a number of distinguished researchers including Prof. Ishiguro from Osaka University attended the seminar. Hido presented the combination of artificial intelligence technologies with applications in automotive and industrial robotics, and showed a demo as an example of new business innovations in the “Socio-economic Impact” session. After the presentation, he was contacted by the audience including British agencies and local consulting firms for future cooperations.

During the stay in London, the Japan group officially visited the newly-established Alan Turing Institute and several start-up companies. Hido also gave a one-hour talk as a public seminar at Imperial College of London and made connections with attendees including faculty members.

Preferred Networks will continue to increase presense not only in the UK but also world-wide for business collaborations and recruitment activities.

TOYOTA and PFN announce capital alliance

PFN is at International Robot Exhibition 2015

Preferred Networks, Inc. is pleased to announce its participation in the International Robot Exhibition 2015, as a part of FANUC’s exhibition, in Tokyo, Japan, December 2nd-5th.

International Robot Exhibition 2015 Official Web
http://biz.nikkan.co.jp/eve/irex/english/

R&D partnership deal with NVIDIA

PFN and NVIDIA agreed on a partnership deal in research and development of deep learning technologies which are used to optimize various industry applications.

Background:
PFN owns leading edge technologies in distributed machine learning, focusing on deep learning, in the areas of industrial robotics, autonomous driving and drug discovery, and is leading the transformation of these industries. These technologies require high-performance data processing capabilities.
NVIDIA has been and will be providing high performance computing technologies accelerated by GPUs and advanced software solutions, in order for data scientists and researchers to maximize their uses of deep learning technologies.

Contents:
PFN has been using NVIDIA’s GPUs for its research and development activities in deep learning. Chainer, an Open-Source Software developed by PFN, uses GPUs to accelerate computing in deep learning modeling processes.
It is necessary to process data on edge devices as well as in the cloud to utilize deep learning technologies in industry segments such as IoT. GPUs play an important role for these edge devices to become more intelligent. We believe this partnership will accelerate the industry trends toward the edge optimization.

Oct22 PFN Career information event

On October 22 (Thursday) Preferred Networks will hold a career information event with talks from the PFN founders Nishikawa and Okanohara, as well as researchers and engineers working at PFN. There will also be a Q&A session where you can get in touch with many of our employees.

Please note: This time, all the talks will be in Japanese, but we are planning to have another event aimed at an English-speaking audience soon. Follow us on Twitter (https://twitter.com/PreferredNet) to stay updated!

Time/Date: Oct 22, 2015, 18:30-20:30 (doors open at 18:00)
Location: エムワイ貸会議室 Ochanomizu
Address: Ochanomizu Union Building (4th floor), 2-1-20 Kandasurugadai, Chiyoda-ku, Tokyo 101-0062

PFN is quickly expanding and we are always looking for great talent. Please visit our job page for more information: https://www.preferred-networks.jp/job_en

Sep7-11 Tobias Pfeiffer Speaks at MobiCom Panel@Paris

Tobias Pfeiffer, Software Engineer at Preferred Networks, participated in a panel at the 21st International Conference on Mobile Computing and Networking (MobiCom), one of the top conferences in the field of mobile computing and wireless networking, that was held from September 7-11 in Paris (France).

In the panel called “Big Data, IoT, … Buzz Words For Academia Or Reality For Industry?”
which was chaired by Wenjun Hu (Yale University, USA), Rui Aguiar (Universidade de Aveiro, Portugal), and Hiroshi Esaki (University of Tokyo, Japan), he presented Preferred Networks’ approach to the Internet of Things, based on Edge-Heavy Computing and Deep Learning. Afterwards, the panelists discussed with the audience the state and future development of the IoT, for example questions such as: “How can academic institutions obtain real-world data for their research?” or “What are important developments that need to happen to advance the current state of IoT?”

FANUC and Preferred Networks announce capital alliance

Tokyo, Japan – FANUC CORPORATION (President: Yoshiharu Inaba; hereinafter, FANUC) and Preferred Networks, Inc. (President and CEO: Toru Nishikawa, hereinafter, PFN) announced today that the two companies have reached an agreement on a capital alliance to collaborate on the technical development of advanced intelligence for industrial machines and robots through application of machine learning and deep learning.

1.Overview of investment by FANUC

Amount of finance: 900 million JPY

Acquired stock: 6.0% of the total number of issued stock of PFN

Method of investment: FANUC underwrites the allocation of new stock of PFN

Date of investment: By the end of September 2015 (Expected)

2.Background of strategic and capital alliance
Intelligent IoT machines will drive the manufacturing industry in its evolution towards Industry 4.0 and the Industrial Internet. However, as machines increase in both intelligence and number, the volume of data being generated also increases, making the problem of how best to utilize this big data more and more important. Advanced automation of industrial machines and robots will be the key to increasing production and efficiency in the future, but it will also generate data that require new methods of analysis. FANUC and PFN have begun joint projects using machine learning and deep learning to analyze such data at the edge of the network—in the machines themselves—in real-time. Today, the two companies announce a capital alliance to accelerate this collaboration.

3.Future direction of collaboration through capital alliance
Until now, deep learning and machine learning have been used primarily for Internet-based applications and services or for retrospective analysis. However, the powers of deep learning and machine learning have not yet been exploited in applications for automated manufacturing processes in the physical world. Through this collaboration, FANUC and PFN aim to integrate PFN’s cutting-edge deep learning and machine learning technologies with FANUC’s industry-leading industrial machines. The focus of the collaboration will be on giving FANUC machines the ability to learn autonomously. For example, FANUC machines will be able to autonomously learn

-complex and non-repetitive behaviors
-strategies and methods for cooperation with other machines
-methods of predicting and repairing malfunctions.

The collaboration will advance the state-of-the-art in the automation of manufacturing processes, including those in Industry 4.0. The two companies plan to work together to innovate throughout the entire range of FANUC products.

[ Contact ]
Preferred Networks, Inc.
pfn-info@preferred.jp

June 17 Announcement for business alliance with Asian Frontier

Announcement of business alliance of Preferred Networks Inc. and Asian Frontier Co., Ltd.
June 17th 2015

Preferred Networks Inc.(PFN) and Asian Frontier Co., Ltd.(Asian Frontier) came to the basic agreement on their business alliance in appointing Asian Frontier as a controlling partner of PFN’s partner companies, who wish to apply PFN’s machine learning and deep learning technologies into their solutions.

[Background]
PFN’s machine learning and deep learning technology have broad deployment potentiality and innovation capability in any business sectors under today’s evolutionary advanced network infrastructure. But, same as many other advanced technologies, deploying such technologies into the business scene, it requires to clarify each business goals and accelerate the development of partner’s business applications properly, which normally took time in the past.
To respond to this issue, PFN appointed Asian Frontier, who has deep experience in Business and IT consultation globally, to efficiently drive the deployment of PFN technology into their partner companies’ business solutions.

[Contents]
PFN plans to establish “Partner Program” for those companies who wish to apply PFN’s machine learning and deep learning technologies into their business solutions.
While PFN keeps focus on the development of his own advanced technology, under this business alliance with Asian Frontier, who brings deep Business and IT consultation know-how, “Partner Program” will benefit partner companies in accelerating their development efficiently.

This “Partner Program” will be started from providing a package software called “DIMo(Deep Intelligence in Motion)”. DIMo provides a platform where each partner company to be able to develop their dedicated solution over the machine learning and deep learning technology.

Asian Frontier, as a controlling partner of this “Partner Program”, will enhance partner companies’ solution development in efficient and timely manner.

* About Asian Frontier
Company Name: Asian Frontier Co., Ltd.
Address: 1-11-28, Nagatacho, Chiyoda-ku, Tokyo, JAPAN
Representative Director: Miguel Angel ESTEVEZ ABE
Founded: September 5th 2007

[Contact]
Preferred Networks Inc.
e-mail: pfn-info@preferred.jp

Asian Frontier Co., Ltd.
http://asian-frontier.com
e-mail:toiawase@asian-frontier.com

June 9-10 Interop Tokyo 2015

Toru Nishikawa, CEO, and Daisuke Okanohara, Executive Vice Present, gave keynote talks at Interop Tokyo 2015 held on Feb 9~10th in Tokyo. On the 10th, PFN announced parthenships with Panasonic and FANUC.

—————
Please refer to the news links below for more details on the partnerships announced.

NEWS
Partnership with Panasonic announcement
Partnership with FANUC announcement

—————
YouTube
A video demonstration of the technology can be viewed at the YouTube link below:
Robot control using Distributed Deep Reinforcement Learning (7:13)
Preferred Research Blog
The technology used in the demo was developed by Eiichi Matsumoto. He explains it in more details at the Research blog below:
Preferred Research Blog (Japanese)

—————
PFN also announced the release of a new deep learning framework “Chainer” as Open Source Software. Seiya Tokui, the Chainer project leader, explains more details about the new framework at the research blog below:

Official site
Chainer.org
NEWS
Release Announcement of “Chainer” : Open Source Deep Learning Framework
Preferred Research Blog
Preferred Research Blog(Japanese)

—————
The announcement were covered in several Japanese and Global Medias:

Wall Street Journal 
http://www.pressreader.com/china/the-wall-street-journal-asia/20150610/281586649216941/TextView
http://www.wsj.com/articles/panasonic-in-deal-to-help-computers-learn-on-the-job-1433840398
http://blogs.wsj.com/cfo/2015/06/10/the-morning-ledger-startups-employ-novel-metrics-to-lure-investment/

The Nikkei Newspaper:(Japanese)
http://www.nikkei.com/article/DGXLASDZ10HRX_Q5A610C1TJC000/
http://www.nikkei.com/article/DGXLASDZ11HJK_R10C15A6TJC000/

The Nikkan Kogyo Newspaper:(Japanese)
http://www.nikkan.co.jp/news/nkx0120150611aaam.html

IT Pro(Japanese)
http://itpro.nikkeibp.co.jp/atcl/news/15/061001951/?top_nhl