Id: arxiv/2601.22370 Url: https://arxiv.org/abs/2601.22370 Tags: math.OC Score: 0 Title: Operator Splitting with Hamilton-Jacobi-based Proximals Domain: math.OC Source: arxiv Authors: Nicholas Di, Eric C. Chi, Samy Wu Fung Reviews: 0 Subject: math.OC Category: research Comments: 0 Disputed: false Division: math Mentions: 0 Discussed: 0 Published: 2026-01-29 Page: https://delphi.science/r/arxiv/2601.22370 Operator splitting algorithms are a cornerstone of modern first-order optimization, decomposing complex problems into simpler subproblems solved via proximal operators. However, most functions lack closed-form proximal operators, which has long restricted these methods to a narrow set of problems. Hamilton-Jacobi-based proximal operator (HJ-Prox) is a recent derivative-free Monte Carlo technique based on Hamilton-Jacobi PDE theory, that approximates proximal operators numerically. In this work, we introduce a unified framework for operator splitting via HJ-Prox, which allows for deployment…