The Power of Learned Locally Linear Models for Nonlinear Policy Optimization
Abstract
A common pipeline in learning-based control is to iteratively estimate a model of system dynamics, and apply a trajectory optimization algorithm - e.g. $\mathtt{iLQR}$ - on the learned model to minimize a target cost. This paper conducts a rigorous analysis of a simplified variant of this strategy for general nonlinear systems. We analyze an algorithm which iterates between estimating local linear models of nonlinear system dynamics and performing $\mathtt{iLQR}$-like policy updates. We demonstrate that this algorithm attains sample complexity polynomial in relevant problem parameters, and, by synthesizing locally stabilizing gains, overcomes exponential dependence in problem horizon. Experimental results validate the performance of our algorithm, and compare to natural deep-learning baselines.
Cite
Text
Pfrommer et al. "The Power of Learned Locally Linear Models for Nonlinear Policy Optimization." International Conference on Machine Learning, 2023.Markdown
[Pfrommer et al. "The Power of Learned Locally Linear Models for Nonlinear Policy Optimization." International Conference on Machine Learning, 2023.](https://mlanthology.org/icml/2023/pfrommer2023icml-power/)BibTeX
@inproceedings{pfrommer2023icml-power,
title = {{The Power of Learned Locally Linear Models for Nonlinear Policy Optimization}},
author = {Pfrommer, Daniel and Simchowitz, Max and Westenbroek, Tyler and Matni, Nikolai and Tu, Stephen},
booktitle = {International Conference on Machine Learning},
year = {2023},
pages = {27737-27821},
volume = {202},
url = {https://mlanthology.org/icml/2023/pfrommer2023icml-power/}
}