Preferred Networks


“Make the real world computable.”

Founded: 2014, Tokyo, Japan

Category: Artificial Intelligence/ Robotics

Primary office: Tokyo, Japan

Core technical team:Tokyo, Japan

Status: Private

Employees: 251 to 500

Amount raised:USD $147.2million (7 rounds – July 2019)


  • Preferred Networks strategizes by applying a combination of deep learning and machine learning to core technologies to develop applications and build large-scale compute clusters providing solutions across a wide variety of domains, including robotics, life science, manufacturing, etc.
  • Distinct competences include Deep learning and Machine learning.
  • Engages in intense Research and Collaboration (R&D)
  • World leader in AI Robotics


  • Valuation: USD $3.3billion (2018)
  • Estimated Revenue: USD $127million (2019)


  • Unveiled a personal robot system – exhibiting fully autonomous tidying-up robots (CEATEC Japan 2018)
  • Most valuable unicorn company (Nikkei Asia 2020)
  • MN- 3 rated world’s most energy efficient supercomputer (Green500 ranking 2020)


  • Chainer Chainer™ – a core deep learning framework. First to adopt the define-by-run approach that allows developers to build complex neural networks in intuitive and flexible ways.
  • CuPy CuPy™ – open-source matrix library accelerated with NVIDIA CUDA.
  • Optuna™ – an open-source automatic hyperparameter optimization framework, automates the trial-and-error process of optimizing the hyperparameters. It automatically finds optimal hyperparameter values based on an optimization target
  • Industrial and Personal Robots
    • Bio & Healthcare – focuses on omics analysis, medical image analysis, and compound analysis using deep learning.
    • Supercomputers – develops chips for artificial intelligence for companies like Google, Huawei Technologies, etc.


  • Partners with Toyota (R&D based Human Support Robot (HSR) robotics platform), Intel (open source framework for deep learning), etc.
  • Open source DevOps community


  • Artificial Intelligence – Machine learning, Deep learning
  • Research & Development
  • Data analytics, Edge-heavy computing, Distributed intelligence.

Distinct AI Features


  • Deep learning, Machine learning

  AI use

  • Robotics
  • Joint research and development of object recognition technologies and vehicle information analysis, which are required for the development of autonomous driving and connected cars.
  • Visual Inspection: proprietary deep learning model offering high accuracy and flexibility for building visual inspection systems.
  • Sports Analytics develops play analytics and pose estimation algorithms for sports using deep learning technologies

AI useRate of return on customer’s investment to make AI work


  • Novel and innovative products as well as solution and/or improvements to business processes.

Long term:

  • Position to be leader in AI-driven solutions for the future


  • Device and Tools-generated data

Quantum Computing

  • N/A



  • Diverse Industry robotic and AI solution projects
  • R&D Collaboration with Toyota, Intel,


  • Research, development & collaborations
  • Open source development


  • Further diversify operations and explore new industries
  • Increase Revenue