ADMM and Accelerated ADMM as Continuous Dynamical Systems
Guilherme Fran\c{c}a, Daniel P. Robinson, Ren\'e Vidal

TL;DR
This paper models ADMM and its accelerated variant as continuous dynamical systems, analyzing their stability and convergence using Lyapunov methods to deepen understanding of their behavior.
Contribution
It derives the differential equations representing the continuous limits of ADMM and accelerated ADMM, providing new insights into their stability and convergence properties.
Findings
Differential equations for ADMM and accelerated ADMM derived
Lyapunov stability analysis applied to these dynamical systems
Convergence rates established for the continuous models
Abstract
Recently, there has been an increasing interest in using tools from dynamical systems to analyze the behavior of simple optimization algorithms such as gradient descent and accelerated variants. This paper strengthens such connections by deriving the differential equations that model the continuous limit of the sequence of iterates generated by the alternating direction method of multipliers, as well as an accelerated variant. We employ the direct method of Lyapunov to analyze the stability of critical points of the dynamical systems and to obtain associated convergence rates.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsDistributed Control Multi-Agent Systems · Stochastic Gradient Optimization Techniques · Model Reduction and Neural Networks
