D-ripALM: A Tuning-friendly Decentralized Relative-Type Inexact Proximal Augmented Lagrangian Method
Abstract
This paper proposes D-ripALM, a Decentralized relative-type inexact proximal Augmented Lagrangian Method for consensus convex optimization over multi-agent networks.
D-ripALM adopts a double-loop distributed optimization framework that accommodates a wide range of inner solvers, enabling efficient treatment of both smooth and nonsmooth objectives.
D-ripALM employs a relative-type error criterion to regulate the switching between inner and outer iterations, replacing a prescribed absolute tolerance sequence with a single relative tolerance parameter.
Moreover, we establish rigorous convergence guarantees for D-ripALM under general convexity assumptions, without requiring smoothness or strong convexity conditions commonly imposed in the distributed optimization literature.
Numerical experiments on synthetic and real datasets demonstrate that a restarted variant of D-ripALM achieves competitive computational efficiency and solution accuracy relative to representative decentralized algorithms.