Designing Neural Network Architectures using Reinforcement Learning

Bowen Baker, Otkrist Gupta, Nikhil Naik, Ramesh Raskar

Nov 04, 2016 (modified: Feb 23, 2017) ICLR 2017 conference submission readers: everyone
  • Abstract: At present, designing convolutional neural network (CNN) architectures requires both human expertise and labor. New architectures are handcrafted by careful experimentation or modified from a handful of existing networks. We introduce MetaQNN, a meta-modeling algorithm based on reinforcement learning to automatically generate high-performing CNN architectures for a given learning task. The learning agent is trained to sequentially choose CNN layers using $Q$-learning with an $\epsilon$-greedy exploration strategy and experience replay. The agent explores a large but finite space of possible architectures and iteratively discovers designs with improved performance on the learning task. On image classification benchmarks, the agent-designed networks (consisting of only standard convolution, pooling, and fully-connected layers) beat existing networks designed with the same layer types and are competitive against the state-of-the-art methods that use more complex layer types. We also outperform existing meta-modeling approaches for network design on image classification tasks.
  • TL;DR: A Q-learning algorithm for automatically generating neural nets
  • Conflicts:,
  • Authorids:,,,
  • Keywords: Deep learning, Reinforcement Learning