OpReg-Boost: Learning to Accelerate Online Algorithms with Operator Regression

Abstract

This paper presents a new regularization approach – termed OpReg-Boost – to boost the convergence of online optimization and learning algorithms. In particular, the paper considers online algorithms for optimization problems with a time-varying (weakly) convex composite cost. For a given online algorithm, OpReg-Boost learns the closest algorithmic map that yields linear convergence; to this end, the learning procedure hinges on the concept of operator regression. We show how to formalize the operator regression problem and propose a computationally-efficient Peaceman-Rachford solver that exploits a closed-form solution of simple quadratically-constrained quadratic programs (QCQPs). Simulation results showcase the superior properties of OpReg-Boost w.r.t. the more classical forward-backward algorithm, FISTA, and Anderson acceleration.

Cite

Text

Bastianello et al. "OpReg-Boost: Learning to Accelerate Online Algorithms with Operator Regression." Proceedings of The 4th Annual Learning for Dynamics and Control Conference, 2022.

Markdown

[Bastianello et al. "OpReg-Boost: Learning to Accelerate Online Algorithms with Operator Regression." Proceedings of The 4th Annual Learning for Dynamics and Control Conference, 2022.](https://mlanthology.org/l4dc/2022/bastianello2022l4dc-opregboost/)

BibTeX

@inproceedings{bastianello2022l4dc-opregboost,
  title     = {{OpReg-Boost: Learning to Accelerate Online Algorithms with Operator Regression}},
  author    = {Bastianello, Nicola and Simonetto, Andrea and Dall’Anese, Emiliano},
  booktitle = {Proceedings of The 4th Annual Learning for Dynamics and Control Conference},
  year      = {2022},
  pages     = {138-152},
  volume    = {168},
  url       = {https://mlanthology.org/l4dc/2022/bastianello2022l4dc-opregboost/}
}