Sketch-and-Project Meets Newton Method: Global $\mathcal O(k^{-2})$ Convergence with Low-Rank Updates

Autor:	Hanzely, Slavomír
Rok vydání:	2023
Předmět:	Mathematics - Optimization and Control Computer Science - Machine Learning
Druh dokumentu:	Working Paper
Popis:	In this paper, we propose the first sketch-and-project Newton method with fast $\mathcal O(k^{-2})$ global convergence rate for self-concordant functions. Our method, SGN, can be viewed in three ways: i) as a sketch-and-project algorithm projecting updates of Newton method, ii) as a cubically regularized Newton ethod in sketched subspaces, and iii) as a damped Newton method in sketched subspaces. SGN inherits best of all three worlds: cheap iteration costs of sketch-and-project methods, state-of-the-art $\mathcal O(k^{-2})$ global convergence rate of full-rank Newton-like methods and the algorithm simplicity of damped Newton methods. Finally, we demonstrate its comparable empirical performance to baseline algorithms. Comment: 10 pages
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2305.13082 Zobrazit plný text záznamu View this record from Arxiv