C'edric F'evotte

On The Linear Convergence of Bregman Proximal Gradient Methods with Applications to Kullback--Leibler regression

This work introduces a novel notion of strong convexity, termed Restricted Relative Strong Convexity, and establishes linear convergence rates for BPGM under this condition, and exploits the proposed theoretical framework to provide an in-depth analysis of the convergence of BPGM for (regularized) Kullback--Leibler regression problems.