Reinforced Genetic Algorithm Learning for Optimizing Computation Graphs
Abstract
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.
Cite
Text
Paliwal et al. "Reinforced Genetic Algorithm Learning for Optimizing Computation Graphs." International Conference on Learning Representations, 2020.Markdown
[Paliwal et al. "Reinforced Genetic Algorithm Learning for Optimizing Computation Graphs." International Conference on Learning Representations, 2020.](https://mlanthology.org/iclr/2020/paliwal2020iclr-reinforced/)BibTeX
@inproceedings{paliwal2020iclr-reinforced,
title = {{Reinforced Genetic Algorithm Learning for Optimizing Computation Graphs}},
author = {Paliwal, Aditya and Gimeno, Felix and Nair, Vinod and Li, Yujia and Lubin, Miles and Kohli, Pushmeet and Vinyals, Oriol},
booktitle = {International Conference on Learning Representations},
year = {2020},
url = {https://mlanthology.org/iclr/2020/paliwal2020iclr-reinforced/}
}