Bowen Baker
bowen@mit.edu
In our paper, Designing Neural Network Architectures Using Reinforcement Learning (arxiv, openreview), we propose a meta-modeling approach 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 ε-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 network design meta-modeling approaches on image classification.
We have just released the full code to run MetaQNN! Find it here.
Bowen Baker
bowen@mit.edu
Otkrist Gupta
otkrist@mit.edu
Nikhil Naik
naik@mit.edu
Ramesh Raskar
raskar@media.mit.edu
Peter Downs
downs@mit.edu