{"id":"arxiv/2601.22682","url":"https://arxiv.org/abs/2601.22682","tags":["math.OC"],"score":0,"title":"SUN-DSBO: A Structured Unified Framework for Nonconvex Decentralized Stochastic Bilevel Optimization","domain":"math.OC","source":"arxiv","authors":["Yaoshuai Ma","Xiao Wang","Wei Yao","Jin Zhang"],"reviews":0,"subject":"math.OC","summary":"Decentralized stochastic bilevel optimization (DSBO) is a powerful tool for various machine learning tasks, including decentralized meta-learning and hyperparameter tuning. Existing DSBO methods primarily address problems with strongly convex lower-level objective functions. However, nonconvex objective functions are increasingly prevalent in modern deep learning. In this work, we introduce SUN-DSBO, a Structured Unified framework for Nonconvex DSBO, in which both the upper- and lower-level objective functions may be nonconvex. Notably, SUN-DSBO offers the flexibility to incorporate…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-01-30","reactions":{}}