Learning to Optimize

Abstract

Algorithm design is a laborious process and often requires many iterations of ideation and validation. In this paper, we explore automating algorithm design and present a method to learn an optimization algorithm. We approach this problem from a reinforcement learning perspective and represent any particular optimization algorithm as a policy. We learn an optimization algorithm using guided policy search and demonstrate that the resulting algorithm outperforms existing hand-engineered algorithms in terms of convergence speed and/or the final objective value.

Cite

Text

Li and Malik. "Learning to Optimize." International Conference on Learning Representations, 2017.

Markdown

[Li and Malik. "Learning to Optimize." International Conference on Learning Representations, 2017.](https://mlanthology.org/iclr/2017/li2017iclr-learning/)

BibTeX

@inproceedings{li2017iclr-learning,
  title     = {{Learning to Optimize}},
  author    = {Li, Ke and Malik, Jitendra},
  booktitle = {International Conference on Learning Representations},
  year      = {2017},
  url       = {https://mlanthology.org/iclr/2017/li2017iclr-learning/}
}