Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.04855 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910862373027840 |
|---|---|
| author | Chen, Yilun Lu, Jiaqi |
| author_facet | Chen, Yilun Lu, Jiaqi |
| contents | We characterize a joint CLT of the number of pulls and the sample mean reward of the arms in a stochastic two-armed bandit environment under UCB algorithms. Several implications of this result are in place: (1) a nonstandard CLT of the number of pulls hence pseudo-regret that smoothly interpolates between a standard form in the large arm gap regime and a slow-concentration form in the small arm gap regime, and (2) a heuristic derivation of the sample bias up to its leading order from the correlation between the number of pulls and sample means. Our analysis framework is based on a novel perturbation analysis, which is of broader interest on its own. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_04855 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A characterization of sample adaptivity in UCB data Chen, Yilun Lu, Jiaqi Machine Learning Probability We characterize a joint CLT of the number of pulls and the sample mean reward of the arms in a stochastic two-armed bandit environment under UCB algorithms. Several implications of this result are in place: (1) a nonstandard CLT of the number of pulls hence pseudo-regret that smoothly interpolates between a standard form in the large arm gap regime and a slow-concentration form in the small arm gap regime, and (2) a heuristic derivation of the sample bias up to its leading order from the correlation between the number of pulls and sample means. Our analysis framework is based on a novel perturbation analysis, which is of broader interest on its own. |
| title | A characterization of sample adaptivity in UCB data |
| topic | Machine Learning Probability |
| url | https://arxiv.org/abs/2503.04855 |