{"id":"arxiv/2603.14762","url":"https://arxiv.org/abs/2603.14762","tags":["math.OC","cs.LG","eess.SY"],"score":0,"title":"Online Learning for Supervisory Switching Control","domain":"math.OC","source":"arxiv","authors":["Haoyuan Sun","Ali Jadbabaie"],"reviews":0,"subject":"math.OC","summary":"We study supervisory switching control for partially-observed linear dynamical systems. The objective is to identify and deploy a suitable controller for the unknown system by periodically selecting among a collection of $N$ candidate controllers, some of which may destabilize the underlying system. While classical estimator-based supervisory control guarantees asymptotic stability, it lacks quantitative finite-time performance bounds. Conversely, current non-asymptotic methods in both online learning and system identification require restrictive assumptions that are incompatible in a…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-03-16","reactions":{}}