{"id":"arxiv/2609.39274","url":"https://arxiv.org/abs/2609.39274","tags":["math.ST","stat.ME"],"score":0,"title":"Adaptive inference for functionals of M-estimands","domain":"math.ST","source":"arxiv","authors":["James Leiner","Aurélien Bibaut","Nathan Kallus","Aaditya Ramdas","Koulik Khamaru"],"reviews":0,"subject":"math.ST","summary":"Reinforcement learning and contextual bandit algorithms have become increasingly common in sequential decision-making applications. When these methods are deployed in high-stakes domains, there is growing interest not only in learning effective policies, but also in conducting statistical inference for quantities learned under adaptive data collection. However, classical procedures applied naively in these settings can fail: even when estimators are unbiased, their variance becomes path-dependent and as a result may not be asymptotically normal. A growing literature has emerged to ameliorate…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-09-30","reactions":{}}