{"id":"arxiv/2605.26985","url":"https://arxiv.org/abs/2605.26985","tags":["math.OC"],"score":0,"title":"A Unified Primal-Dual Recipe for Accelerating Three-Operator Splitting Methods","domain":"math.OC","source":"arxiv","authors":["Abdurakhmon Sadiev","Laurent Condat","Peter Richtárik"],"reviews":0,"subject":"math.OC","summary":"Composite optimization problems, formulated as the minimization of three functions, are ubiquitous in large-scale machine learning and signal processing. While state-of-the-art splitting methods such as Condat-Vũ (CV) [Condat, 2013, Vũ, 2013], Primal-Dual Davis-Yin (PDDY) [Salim et al., 2022b], and Primal-Dual Twice Reflected (PDTR) [Malitsky and Tam, 2026] are highly versatile, they inherently exhibit non-accelerated convergence rates. Existing accelerated primal-dual splitting results either focus on special structures like one of the functions being zero, or linearly constrained problems,…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-05-26","reactions":{}}