We present a deep reinforcement learning approach to minimizing the execution cost of neural network computation graphs in an optimizing compiler. Unlike earlier learning-based works that require training the optimizer on the same graph to be optimized, we propose a learning approach that trains an optimizer offline and then generalizes to previously unseen graphs without further training. This allows our approach to produce high-quality execution decisions on real-world TensorFlow graphs in seconds instead of hours. We consider two optimization tasks for computation graphs: minimizing running time and peak memory usage. In comparison to an extensive set of baselines, our approach achieves significant improvements over classical and other learning-based methods on these two tasks.
@article{arxiv.1905.02494,
title = {Reinforced Genetic Algorithm Learning for Optimizing Computation Graphs},
author = {Aditya Paliwal and Felix Gimeno and Vinod Nair and Yujia Li and Miles Lubin and Pushmeet Kohli and Oriol Vinyals},
journal= {arXiv preprint arXiv:1905.02494},
year = {2020}
}
Comments
Accepted to ICLR 2020 https://openreview.net/forum?id=rkxDoJBYPB