Bayesian Best-Arm Identification with Abstention: A Polynomial-to-Exponential Phase Transition

• URL: http://arxiv.org/abs/2606.29203v1
• Date: Sun, 28 Jun 2026 05:07:27 GMT
• Title: Bayesian Best-Arm Identification with Abstention: A Polynomial-to-Exponential Phase Transition
• Abstract summary: We study the Bayesian fixed-budget best-arm identification problem in which a learner can abstain from making a terminal recommendation.<n>We analyze the probability of undetected error--the risk of recommending a suboptimal arm without abstaining.
• Score: 51.253617466579335
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We study the Bayesian fixed-budget best-arm identification problem in which a learner can abstain from making a terminal recommendation. Subject to an abstention budget $α$, we analyze the probability of undetected error--the risk of recommending a suboptimal arm without abstaining. Our central finding is that abstention induces a phase transition: without abstention, the error probability decays polynomially in the sampling budget $T$; in contrast, introducing any small positive abstention budget shifts this to an exponential decay. For Gaussian priors and rewards, in the regime $T\to\infty$ followed by $α\downarrow0$, we establish exact matching information-theoretic lower bounds and algorithmic upper bounds on the optimal error exponent, which takes the form $\exp(-\frac{α^{2}T}{8κ_ν^{2}})$. The hardness parameter $κ_ν$ represents the prior density of the top-two gap at zero, highlighting that nearly tied instances drive the fundamental error. We introduce an adaptive algorithm, PGWS, that successfully achieves this optimal exponent by expending its abstention budget on statistically ambiguous instances. We further demonstrate that this polynomial-to-exponential improvement is exclusively a Bayesian phenomenon--in the frequentist setting, abstention only affects lower-order exponent terms. We also extend our results beyond the Gaussian model.

Related papers

• Does 1/2-Tsallis-INF Also Work Well for Best-Arm Identification? [11.223878908981725]
  $1/2$-Tsallis-INF is a canonical best-of-both-worlds FTRL algorithm.<n>It achieves logarithmic pseudo-regret in bandits while retaining minimax-optimal regret in adversarial bandits.<n>This raises a natural question: can the same algorithm, without additional exploration, also identify the best arm reliably?
  arXiv  Detail & Related papers  (2026-08-15T18:38:36Z)
• Adaptive Confidence Intervals in Efron's Gaussian Two-Groups Model [28.636700996382558]
  Motivated by noise-oblivious nature of adversaries, we study confidence intervals for the null location parameter $$ in Efron's Gaussian two-groups fraction model.<n>We characterize the minimax-optimal length among confidence intervals with a prescribed coverage level uniformly over the unknown proportion contamination and all noise-oblivious adversaries.
  arXiv  Detail & Related papers  (2026-04-28T22:40:54Z)
• Provably Adaptive Linear Approximation for the Shapley Value and Beyond [73.0940890296463]
  A fundamental and long-standing challenge is their efficient approximation.<n>We develop a linear-space algorithm that requires $P(|hatboldsymbol-boldsymbol|_2geq)leq $ for all commonly used semi-values.<n>Our algorithm allows explicit minimization of the mean square error for each specific utility function.
  arXiv  Detail & Related papers  (2026-04-09T16:38:14Z)
• Asymptotically Optimal Linear Best Feasible Arm Identification with Fixed Budget [55.938644481736446]
  We introduce a novel algorithm for best feasible arm identification that guarantees an exponential decay in the error probability.<n>We validate our algorithm through comprehensive empirical evaluations across various problem instances with different levels of complexity.
  arXiv  Detail & Related papers  (2025-06-03T02:56:26Z)
• Continuous K-Max Bandits [54.21533414838677]
  We study the $K$-Max multi-armed bandits problem with continuous outcome distributions and weak value-index feedback.<n>This setting captures critical applications in recommendation systems, distributed computing, server scheduling, etc.<n>Our key contribution is the computationally efficient algorithm DCK-UCB, which combines adaptive discretization with bias-corrected confidence bounds.
  arXiv  Detail & Related papers  (2025-02-19T06:37:37Z)
• Sign Operator for Coping with Heavy-Tailed Noise in Non-Convex Optimization: High Probability Bounds Under $(L_0, L_1)$-Smoothness [74.18546828528298]
  We show that SignSGD with Majority Voting can robustly work on the whole range of complexity with $kappakappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappakappakappa-1right, kappa
  arXiv  Detail & Related papers  (2025-02-11T19:54:11Z)
• Fixed-Budget Differentially Private Best Arm Identification [62.36929749450298]
  We study best arm identification (BAI) in linear bandits in the fixed-budget regime under differential privacy constraints. We derive a minimax lower bound on the error probability, and demonstrate that the lower and the upper bounds decay exponentially in $T$.
  arXiv  Detail & Related papers  (2024-01-17T09:23:25Z)
• An $\varepsilon$-Best-Arm Identification Algorithm for Fixed-Confidence and Beyond [15.871535287520393]
  We propose EB-TC$varepsilon$, a novel sampling rule for best arm identification in bandits. We provide three types of theoretical guarantees for EB-TC$varepsilon$. We show through numerical simulations that EB-TC$varepsilon$ performs favorably compared to existing algorithms.
  arXiv  Detail & Related papers  (2023-05-25T13:19:11Z)
• Suboptimal Performance of the Bayes Optimal Algorithm in Frequentist Best Arm Identification [5.176134438571082]
  We consider the fixed-budget best arm identification problem with rewards following normal distributions. In this problem, the forecaster is given $K$ arms (or treatments) and $T$ time steps. The algorithm's performance is evaluated by simple regret, reflecting the quality of the estimated best arm.
  arXiv  Detail & Related papers  (2022-02-10T17:50:26Z)
• Optimal Regret Is Achievable with Bounded Approximate Inference Error: An Enhanced Bayesian Upper Confidence Bound Framework [22.846260353176614]
  We propose an Enhanced Bayesian Upper Confidence Bound (EBUCB) framework that can efficiently accommodate bandit problems. We show that EBUCB can achieve the optimal regret order $O(log T)$ if the inference error measured by two different $alpha$-divergences is less than a constant. Our study provides the first theoretical regret bound that is better than $o(T)$ in the setting of constant approximate inference error.
  arXiv  Detail & Related papers  (2022-01-31T01:36:15Z)
• Learning Halfspaces with Tsybakov Noise [50.659479930171585]
  We study the learnability of halfspaces in the presence of Tsybakov noise. We give an algorithm that achieves misclassification error $epsilon$ with respect to the true halfspace.
  arXiv  Detail & Related papers  (2020-06-11T14:25:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.