{"id":"arxiv/2607.10791","url":"https://arxiv.org/abs/2607.10791","tags":["math.OC"],"score":0,"title":"Resilient Decentralized Preconditioned Stochastic Proximal Optimization with Gradient Sketching over Time-Varying Networks","domain":"math.OC","source":"arxiv","authors":["Jinhui Hu","Guo Chen","Huaqing Li","Tao Han","Tingwen Huang"],"reviews":0,"subject":"math.OC","summary":"Variance reduction is indispensable in Byzantine-resilient decentralized stochastic optimization over multi-agent systems (MASs), as it effectively mitigates gradient noise to facilitate the resilient aggregation process. However, most existing Byzantine-resilient decentralized variance-reduced (VR) stochastic gradient algorithms rely on random data sampling, which proves inefficient in data-scarce yet high-dimensional tasks, for instance, image deblurring. This paper aims to pursue an alternative approach that achieves variance reduction via gradient sketching. We first formulate a class of…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-07-12","reactions":{}}