September 26-27, 2016
New York, NY

Chainer: A flexible and intuitive framework for complex neural networks

Orion Wolfe (Preferred Networks), Shohei Hido (Preferred Networks)
3:45pm–4:25pm Tuesday, 09/27/2016
Location: River Pavilion B
Average rating: ****.
(4.00, 1 rating)

What you'll learn

  • Explore Chainer, a Python-based standalone framework that enables users to intuitively implement many kinds of other models, including recurrent neural networks, with a lot of flexibility and comparable performance to GPUs
  • Description

    Open source software frameworks are the key for applying deep learning technologies. Due to the success of Caffe, Torch, Theano, and TensorFlow, the power of deep learning continues to expand beyond traditional pattern recognition tasks such as image recognition. However, the gap is rapidly increasing between the complexities of newly proposed neural network models and the capabilities of existing frameworks, which have been mainly used for convolutional neural networks. Orion Wolfe and Shohei Hido introduce Chainer, a Python-based standalone framework that enables users to intuitively implement many kinds of other models, including recurrent neural networks, with a lot of flexibility and comparable performance to GPUs.

    Photo of Orion Wolfe

    Orion Wolfe

    Preferred Networks

    Orion Wolfe started his career working on aerospace guidance, controls, modeling, and simulation. He then worked as quantitative trader, developing strategies as part of a statistical arbitrage trading group. Inspired to learn more about computer science while still furthering his skills in modeling, Orion joined a research group in applied machine learning with a focus on semiconductor manufacturing tools and anomaly detection. He holds a BS with honors in electrical engineering from UCLA and an MS in management science and engineering from Stanford. While in college, Orion performed research in particle physics and support vector machines.

    Photo of Shohei Hido

    Shohei Hido

    Preferred Networks

    Shohei Hido is the chief research officer of Preferred Networks, a spin-off company of Preferred Infrastructure, Inc., where he is currently responsible for Deep Intelligence in Motion, a software platform for using deep learning in IoT applications. Previously, Shohei was the leader of Preferred Infrastructure’s Jubatus project, an open source software framework for real-time, streaming machine learning and worked at IBM Research in Tokyo for six years as a staff researcher in machine learning and its applications to industries. Shohei holds an MS in informatics from Kyoto University.