When May a Bandit Leave Its Anchor? E-Process-Authorized Thompson Sampling under Non-stationarity
Abstract
Stationarity rewards memory, but after a change the same history can mislead. We ask when forgetting should be permitted. E-process-authorized Thompson sampling (e-ATS) gives each arm full-history and discounted Beta states. An anytime-valid e-process first authorizes the discounted state, then a reversible relevance score controls its influence. Before authorization, e-ATS exactly follows optimistic Thompson sampling (OTS). Under a Beta-Bernoulli prior-predictive stationary model, e-ATS's probability of ever departing from OTS is at most the chosen αE, without fitted thresholds. Relative to e-ATS, removing authorization increased mean normalized dynamic pseudo-regret by 38.4% on the registered suite but reduced it by 7.5% on the literature-derived replay suite. Therefore, evidence controls when adaptation begins, not whether it always helps.
Related
- Convergence Rates of Gaussian Process Thompson Sampling for Efficient Global Optimization
- Improved Regret Analysis for Parallel Gaussian Process Bandit Optimization
- Exploration strategies for Gaussian process bandit optimization with inference regret guarantees
- Bandit Process Inference for Streaming Functional Data
- Thompson Sampling for Learning in Zero-Sum Games
BibTeX
@misc{gulati2026when,
title = {{When May a Bandit Leave Its Anchor? E-Process-Authorized Thompson Sampling under Non-stationarity}},
author = {Mayand Gulati and Kerong Wang and WeiChen Au},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.03646},
eprint = {2610.03646},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.03646},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - When May a Bandit Leave Its Anchor? E-Process-Authorized Thompson Sampling under Non-stationarity
AU - Mayand Gulati
AU - Kerong Wang
AU - WeiChen Au
PY - 2026
DA - 2026/10/02
PB - arXiv
DO - 10.48550/arxiv.2610.03646
UR - https://arxiv.org/abs/2610.03646
AB - Stationarity rewards memory, but after a change the same history can mislead. We ask when forgetting should be permitted. E-process-authorized Thompson sampling (e-ATS) gives each arm full-history and discounted Beta states. An anytime-valid e-process first authorizes the discounted state, then a reversible relevance score controls its influence. Before authorization, e-ATS exactly follows optimistic Thompson sampling (OTS). Under a Beta-Bernoulli prior-predictive stationary model, e-ATS's probability of ever departing from OTS is at most the chosen α_E, without fitted thresholds. Relative to e-ATS, removing authorization increased mean normalized dynamic pseudo-regret by 38.4% on the registered suite but reduced it by 7.5% on the literature-derived replay suite. Therefore, evidence controls when adaptation begins, not whether it always helps.
ER -