Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.07208 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913164855083008 |
|---|---|
| author | Qu, Yanlin Namkoong, Hongseok Zeevi, Assaf |
| author_facet | Qu, Yanlin Namkoong, Hongseok Zeevi, Assaf |
| contents | Thompson Sampling is one of the most widely used and studied bandit algorithms, known for its simple structure, low regret performance, and solid theoretical guarantees. Yet, in stark contrast to most other families of bandit algorithms, the exact mechanism through which posterior sampling (as introduced by Thompson) is able to "properly" balance exploration and exploitation, remains a mystery. In this paper, we show that the core insight to address this question stems from recasting Thompson Sampling as an online optimization algorithm. To distill this, we introduce a suitable time invariant notion of regret that leads to a stationarized bandit problem, and a stationary Bellman-optimal policy. We then show that Thompson Sampling admits an online optimization form that mimics the structure of the aforementioned Bellman-optimal policy, where "greediness" is regularized by a measure of residual uncertainty. This new lens of online optimization allows both a better understanding of Thompson Sampling dynamics, as well as a principled manner for policy improvement that mimics the Bellman-optimal benchmark. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_07208 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Broader View of Thompson Sampling Qu, Yanlin Namkoong, Hongseok Zeevi, Assaf Machine Learning Thompson Sampling is one of the most widely used and studied bandit algorithms, known for its simple structure, low regret performance, and solid theoretical guarantees. Yet, in stark contrast to most other families of bandit algorithms, the exact mechanism through which posterior sampling (as introduced by Thompson) is able to "properly" balance exploration and exploitation, remains a mystery. In this paper, we show that the core insight to address this question stems from recasting Thompson Sampling as an online optimization algorithm. To distill this, we introduce a suitable time invariant notion of regret that leads to a stationarized bandit problem, and a stationary Bellman-optimal policy. We then show that Thompson Sampling admits an online optimization form that mimics the structure of the aforementioned Bellman-optimal policy, where "greediness" is regularized by a measure of residual uncertainty. This new lens of online optimization allows both a better understanding of Thompson Sampling dynamics, as well as a principled manner for policy improvement that mimics the Bellman-optimal benchmark. |
| title | A Broader View of Thompson Sampling |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2510.07208 |