{"id":"arxiv/2601.21131","url":"https://arxiv.org/abs/2601.21131","tags":["math.ST","cs.IT","cs.LG","stat.ML"],"score":0,"title":"Thompson sampling: Precise arm-pull dynamics and adaptive inference","domain":"math.ST","source":"arxiv","authors":["Qiyang Han"],"reviews":0,"subject":"math.ST","summary":"Adaptive sampling schemes are well known to create complex dependence that may invalidate conventional inference methods. A recent line of work shows that this need not be the case for UCB-type algorithms in multi-armed bandits. A central emerging theme is a `stability' property with asymptotically deterministic arm-pull counts in these algorithms, making inference as easy as in the i.i.d. setting. In this paper, we study the precise arm-pull dynamics in another canonical class of Thompson-sampling type algorithms. We show that the phenomenology is qualitatively different: the arm-pull count…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-01-29","reactions":{}}