Learning-to-learn non-convex piecewise-Lipschitz functions
Abstract
We analyze the meta-learning of the initialization and step-size of learning algorithms for piecewise-Lipschitz functions, a non-convex setting with applications to both machine learning and algorithms. Starting from recent regret bounds for the exponential forecaster on losses with dispersed discontinuities, we generalize them to be initialization-dependent and then use this result to propose a practical meta-learning procedure that learns both the initialization and the step-size of the algorithm from multiple online learning tasks. Asymptotically, we guarantee that the average regret across tasks scales with a natural notion of task-similarity that measures the amount of overlap between near-optimal regions of different tasks. Finally, we instantiate the method and its guarantee in two important settings: robust meta-learning and multi-task data-driven algorithm design.
Cite
@article{arxiv.2108.08770,
title = {Learning-to-learn non-convex piecewise-Lipschitz functions},
author = {Maria-Florina Balcan and Mikhail Khodak and Dravyansh Sharma and Ameet Talwalkar},
journal= {arXiv preprint arXiv:2108.08770},
year = {2021}
}