A systematic approach to Lyapunov analyses of continuous-time models in convex optimization

05/25/2022
by   Céline Moucer, et al.
0

First-order methods are often analyzed via their continuous-time models, where their worst-case convergence properties are usually approached via Lyapunov functions. In this work, we provide a systematic and principled approach to find and verify Lyapunov functions for classes of ordinary and stochastic differential equations. More precisely, we extend the performance estimation framework, originally proposed by Drori and Teboulle [10], to continuous-time models. We retrieve convergence results comparable to those of discrete methods using fewer assumptions and convexity inequalities, and provide new results for stochastic accelerated gradient flows.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset