Learning-to-Learn to Guide Random Search: Derivative-Free Meta Blackbox Optimization on Manifold

Abstract

Solving a sequence of high-dimensional, nonconvex, but potentially similar optimization problems poses a computational challenge in engineering applications. We propose the first meta-learning framework that leverages the shared structure among sequential tasks to improve the computational efficiency and sample complexity of derivative-free optimization. Based on the observation that most practical high-dimensional functions lie on a latent low-dimensional manifold, which can be further shared among instances, our method jointly learns the meta-initialization of a search point and a meta-manifold. Theoretically, we establish the benefit of meta-learning in this challenging setting. Empirically, we demonstrate the effectiveness of the proposed algorithm in two high-dimensional reinforcement learning tasks.

Cite

Text

Sel et al. "Learning-to-Learn to Guide Random Search: Derivative-Free Meta Blackbox Optimization on Manifold." Proceedings of The 5th Annual Learning for Dynamics and Control Conference, 2023.

Markdown

[Sel et al. "Learning-to-Learn to Guide Random Search: Derivative-Free Meta Blackbox Optimization on Manifold." Proceedings of The 5th Annual Learning for Dynamics and Control Conference, 2023.](https://mlanthology.org/l4dc/2023/sel2023l4dc-learningtolearn/)

BibTeX

@inproceedings{sel2023l4dc-learningtolearn,
  title     = {{Learning-to-Learn to Guide Random Search: Derivative-Free Meta Blackbox Optimization on Manifold}},
  author    = {Sel, Bilgehan and Tawaha, Ahmad and Ding, Yuhao and Jia, Ruoxi and Ji, Bo and Lavaei, Javad and Jin, Ming},
  booktitle = {Proceedings of The 5th Annual Learning for Dynamics and Control Conference},
  year      = {2023},
  pages     = {38-50},
  volume    = {211},
  url       = {https://mlanthology.org/l4dc/2023/sel2023l4dc-learningtolearn/}
}