Potential-Function Proofs for First-Order Methods
Machine Learning
2019-06-04 v3 Data Structures and Algorithms
Optimization and Control
Abstract
This note discusses proofs for convergence of first-order methods based on simple potential-function arguments. We cover methods like gradient descent (for both smooth and non-smooth settings), mirror descent, and some accelerated variants.
Cite
@article{arxiv.1712.04581,
title = {Potential-Function Proofs for First-Order Methods},
author = {Nikhil Bansal and Anupam Gupta},
journal= {arXiv preprint arXiv:1712.04581},
year = {2019}
}