{"id":"arxiv/2602.16148","url":"https://arxiv.org/abs/2602.16148","tags":["math.OC","cs.LG"],"score":0,"title":"Local adapt-then-combine algorithms for distributed nonsmooth optimization: Achieving provable communication acceleration","domain":"math.OC","source":"arxiv","authors":["Luyao Guo","Xinli Shi","Wenying Xu","Jinde Cao"],"reviews":0,"subject":"math.OC","summary":"This paper is concerned with the distributed composite optimization problem over networks, where agents aim to minimize a sum of local smooth components and a common nonsmooth term. Leveraging the probabilistic local updates mechanism, we propose a communication-efficient Adapt-Then-Combine (ATC) framework, FlexATC, unifying numerous ATC-based distributed algorithms. Under stepsizes independent of the network topology and the number of local updates, we establish sublinear and linear convergence rates for FlexATC in convex and strongly convex settings, respectively. Remarkably, in the strong…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-02-18","reactions":{}}