aiqbench analysis
This paper presents a theoretical advancement in Thompson Sampling by introducing fractional posteriors ($\alpha$-TS) to improve regret bounds for stochastic multi-armed bandit problems. The analysis demonstrates that tempering the likelihood with a factor $\alpha \in (0,1)$ yields tighter instance-dependent guarantees under mild conditions without requiring conjugate priors or closed-form posteriors. These findings offer valuable insights for researchers optimizing exploration-exploitation strategies in reinforcement learning and sequential decision-making systems.
Generalized Regret Analysis of Thompson Sampling using Fractional Posteriors
arXiv:2309.06349v2 Announce Type: replace-cross Abstract: Thompson sampling (TS) is one of the most popular and earliest algorithms to solve stochastic multi-armed bandit problems. We consider a variant of TS, named $\alpha$-TS, where we use a fractional or $\alpha$-posterior ($\alpha\in(0,1)$) instead of the standard posterior distribution. To compute an $\alpha$-posterior, the likelihood in the definition of the standard posterior is tempered with a factor $\alpha$. For $\alpha$-TS we obtain both instance-dependent $\mathcal{O}\left(\sum_{k \neq i^*} \Delta_k\left(\frac{\log(T)}{C(\alpha)\Delta_k^2} + \frac{1}{2} \right)\right)$ and instance-independent $\mathcal{O}(\sqrt{KT\log K})$ frequentist regret bounds under very mild conditions on the prior and reward distributions, where $\Delta_k$ is the gap between the true mean rewards of the $k^{th}$ and the best arms, and $C(\alpha)$ is a known constant. Both the sub-Gaussian and exponential family models satisfy our general conditions on the reward distribution. Our conditions on the prior distribution can be easily satisfied by a density that is positive, continuous, and bounded. We also establish another instance-dependent regret upper bound that matches (up to constants) to that of improved UCB [Auer and Ortner, 2010]. Our regret analysis carefully adapts and combines recent theoretical developments in the non-asymptotic concentration analysis and Bernstein-von Mises type results for the $\alpha$-posterior distribution. Moreover, our analysis does not require additional structural properties such as closed-form posteriors or conjugate priors.
Abstract:Thompson sampling (TS) is one of the most popular and earliest algorithms to solve stochastic multi-armed bandit problems. We consider a variant of TS, named $\alpha$-TS, where we use a fractional or $\alpha$-posterior ($\alpha\in(0,1)$) instead of the standard posterior distribution. To compute an $\alpha$-posterior, the likelihood in the definition of the standard posterior is tempered with a factor $\alpha$. For $\alpha$-TS we obtain both instance-dependent $\mathcal{O}\left(\sum_{k \neq i^*} \Delta_k\left(\frac{\log(T)}{C(\alpha)\Delta_k^2} + \frac{1}{2} \right)\right)$ and instance-independent $\mathcal{O}(\sqrt{KT\log K})$ frequentist regret bounds under very mild conditions on the prior and reward distributions, where $\Delta_k$ is the gap between the true mean rewards of the $k^{th}$ and the best arms, and $C(\alpha)$ is a known constant. Both the sub-Gaussian and exponential family models satisfy our general conditions on the reward distribution. Our conditions on the prior distribution can be easily satisfied by a density that is positive, continuous, and bounded. We also establish another instance-dependent regret upper bound that matches (up to constants) to that of improved UCB [Auer and Ortner, 2010]. Our regret analysis carefully adapts and combines recent theoretical developments in the non-asymptotic concentration analysis and Bernstein-von Mises type results for the $\alpha$-posterior distribution. Moreover, our analysis does not require additional structural properties such as closed-form posteriors or conjugate priors.
Machine Learning (stat.ML); Machine Learning (cs.LG); Systems and Control (eess.SY); Optimization and Control (math.OC); Statistics Theory (math.ST)
arXiv:2309.06349 [stat.ML]
(or arXiv:2309.06349v2 [stat.ML] for this version)
https://doi.org/10.48550/arXiv.2309.06349 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 314 words
Related coverage
Posterior Tempering Explains Variance Inflation in Linear and Generalized Linear Thompson Sampling
arXiv · Sep 3, 2026
Dynamic Decision-Making under Model Misspecification: A Stochastic Stability Approach
arXiv · Jul 21, 2026
Sharp Deviations Bounds for Dirichlet Weighted Sums with Application to analysis of Bayesian algorithms
arXiv · Oct 6, 2026
Exact Bayes Regret and Asymptotic Optimality in High-Dimensional Gaussian Bandits
arXiv · Sep 26, 2026