Regret Distribution in Stochastic Bandits: Optimal Trade-off between Expectation and Tail Risk

• URL: http://arxiv.org/abs/2304.04341v2
• Date: Fri, 24 Oct 2025 01:31:21 GMT
• Title: Regret Distribution in Stochastic Bandits: Optimal Trade-off between Expectation and Tail Risk
• Authors: David Simchi-Levi, Zeyu Zheng, Feng Zhu,
• Abstract summary: We study the optimal trade-off between expectation and tail risk for regret distribution in the multi-armed bandit model.<n>New policies are proposed to characterize the optimal regret tail probability for any regret threshold.
• Score: 26.397343668067382
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We study the optimal trade-off between expectation and tail risk for regret distribution in the stochastic multi-armed bandit model. We fully characterize the interplay among three desired properties for policy design: worst-case optimality, instance-dependent consistency, and light-tailed risk. New policies are proposed to characterize the optimal regret tail probability for any regret threshold. In particular, we discover an intrinsic gap of the optimal tail rate depending on whether the time horizon $T$ is known a priori or not. Interestingly, when it comes to the purely worst-case scenario, this gap disappears. Our results reveal insights on how to design policies that balance between efficiency and safety, and highlight extra insights on policy robustness with regard to policy hyper-parameters and model mis-specification. We also conduct a simulation study to validate our theoretical insights and provide practical amendment to our policies. Finally, we discuss extensions of our results to (i) general sub-exponential environments and (ii) general stochastic linear bandits. Furthermore, we find that a special case of our policy design surprisingly coincides with what was adopted in AlphaGo Monte Carlo Tree Search. Our theory provides high-level insights to why their engineered solution is successful and should be advocated in complex decision-making environments.

Related papers

• Learning Optimal Deterministic Policies with Stochastic Policy Gradients [62.81324245896716]
  Policy gradient (PG) methods are successful approaches to deal with continuous reinforcement learning (RL) problems. In common practice, convergence (hyper)policies are learned only to deploy their deterministic version. We show how to tune the exploration level used for learning to optimize the trade-off between the sample complexity and the performance of the deployed deterministic policy.
  arXiv  Detail & Related papers  (2024-05-03T16:45:15Z)
• Model-Based Epistemic Variance of Values for Risk-Aware Policy Optimization [59.758009422067]
  We consider the problem of quantifying uncertainty over expected cumulative rewards in model-based reinforcement learning. We propose a new uncertainty Bellman equation (UBE) whose solution converges to the true posterior variance over values. We introduce a general-purpose policy optimization algorithm, Q-Uncertainty Soft Actor-Critic (QU-SAC) that can be applied for either risk-seeking or risk-averse policy optimization.
  arXiv  Detail & Related papers  (2023-12-07T15:55:58Z)
• Risk-sensitive Markov Decision Process and Learning under General Utility Functions [3.069335774032178]
  Reinforcement Learning (RL) has gained substantial attention across diverse application domains and theoretical investigations.<n>We consider a scenario where the decision-maker seeks to optimize a general utility function of the cumulative reward in the framework of a decision process (MDP)<n>We propose a modified value iteration algorithm that employs an epsilon-covering over the space of cumulative reward.<n>In the absence of a simulator, our algorithm, designed with an upper-confidence-bound exploration approach, identifies a near-optimal policy.
  arXiv  Detail & Related papers  (2023-11-22T18:50:06Z)
• $(\epsilon, u)$-Adaptive Regret Minimization in Heavy-Tailed Bandits [29.966828248335972]
  We study the regret minimization problem when $epsilon$ and $u$ are unknown to the learner. AdaR-UCB is the first algorithm that enjoys regret guarantees nearly matching those of the non-adaptive heavy-tailed case.
  arXiv  Detail & Related papers  (2023-10-04T17:11:15Z)
• Probably Anytime-Safe Stochastic Combinatorial Semi-Bandits [81.60136088841948]
  We propose an algorithm that minimizes the regret over the horizon of time $T$. The proposed algorithm is applicable to domains such as recommendation systems and transportation.
  arXiv  Detail & Related papers  (2023-01-31T03:49:00Z)
• Autoregressive Bandits [58.46584210388307]
  We propose a novel online learning setting, Autoregressive Bandits, in which the observed reward is governed by an autoregressive process of order $k$. We show that, under mild assumptions on the reward process, the optimal policy can be conveniently computed. We then devise a new optimistic regret minimization algorithm, namely, AutoRegressive Upper Confidence Bound (AR-UCB), that suffers sublinear regret of order $widetildemathcalO left( frac(k+1)3/2sqrtnT (1-G
  arXiv  Detail & Related papers  (2022-12-12T21:37:36Z)
• A Simple and Optimal Policy Design with Safety against Heavy-Tailed Risk for Stochastic Bandits [27.058087400790555]
  We study the multi-armed bandit problem and design new policies that enjoy both worst-case optimality for expected regret and light-tailed risk for regret distribution. We find that from a managerial perspective, our new policy design yields better tail distributions and is preferable than celebrated policies.
  arXiv  Detail & Related papers  (2022-06-07T02:10:30Z)
• Policy Gradient Bayesian Robust Optimization for Imitation Learning [49.881386773269746]
  We derive a novel policy gradient-style robust optimization approach, PG-BROIL, to balance expected performance and risk. Results suggest PG-BROIL can produce a family of behaviors ranging from risk-neutral to risk-averse.
  arXiv  Detail & Related papers  (2021-06-11T16:49:15Z)
• Stochastic Shortest Path: Minimax, Parameter-Free and Towards Horizon-Free Regret [144.6358229217845]
  We study the problem of learning in the shortest path (SSP) setting, where an agent seeks to minimize the expected cost accumulated before reaching a goal state. We design a novel model-based algorithm EB-SSP that carefully skews the empirical transitions and perturbs the empirical costs with an exploration bonus. We prove that EB-SSP achieves the minimax regret rate $widetildeO(B_star sqrtS A K)$, where $K$ is the number of episodes, $S$ is the number of states, $A$
  arXiv  Detail & Related papers  (2021-04-22T17:20:48Z)
• Stability and Deviation Optimal Risk Bounds with Convergence Rate $O(1/n)$ [4.1499725848998965]
  We show a high probability excess risk bound with the rate $O(log n/n)$ for strongly convex and Lipschitz losses valid for emphany empirical risk minimization method. We discuss how $O(log n/n)$ high probability excess risk bounds are possible for projected gradient descent in the case of strongly convex and Lipschitz losses without the usual smoothness assumption.
  arXiv  Detail & Related papers  (2021-03-22T17:28:40Z)
• Fast Rates for the Regret of Offline Reinforcement Learning [69.23654172273085]
  We study the regret of reinforcement learning from offline data generated by a fixed behavior policy in an infinitehorizon discounted decision process (MDP) We show that given any estimate for the optimal quality function $Q*$, the regret of the policy it defines converges at a rate given by the exponentiation of the $Q*$-estimate's pointwise convergence rate.
  arXiv  Detail & Related papers  (2021-01-31T16:17:56Z)
• Minimax Off-Policy Evaluation for Multi-Armed Bandits [58.7013651350436]
  We study the problem of off-policy evaluation in the multi-armed bandit model with bounded rewards. We develop minimax rate-optimal procedures under three settings.
  arXiv  Detail & Related papers  (2021-01-19T18:55:29Z)
• Off-Policy Evaluation of Slate Policies under Bayes Risk [70.10677881866047]
  We study the problem of off-policy evaluation for slate bandits, for the typical case in which the logging policy factorizes over the slots of the slate. We show that the risk improvement over PI grows linearly with the number of slots, and linearly with the gap between the arithmetic and the harmonic mean of a set of slot-level divergences.
  arXiv  Detail & Related papers  (2021-01-05T20:07:56Z)
• Risk-Sensitive Deep RL: Variance-Constrained Actor-Critic Provably Finds Globally Optimal Policy [95.98698822755227]
  We make the first attempt to study risk-sensitive deep reinforcement learning under the average reward setting with the variance risk criteria. We propose an actor-critic algorithm that iteratively and efficiently updates the policy, the Lagrange multiplier, and the Fenchel dual variable.
  arXiv  Detail & Related papers  (2020-12-28T05:02:26Z)
• Optimal Algorithms for Stochastic Multi-Armed Bandits with Heavy Tailed Rewards [24.983866845065926]
  We consider multi-armed bandits with heavy-tailed rewards, whose $p$-th moment is bounded by a constant $nu_p$ for $1pleq2$. We propose a novel robust estimator which does not require $nu_p$ as prior information. We show that an error probability of the proposed estimator decays exponentially fast.
  arXiv  Detail & Related papers  (2020-10-24T10:44:02Z)
• Bayesian Robust Optimization for Imitation Learning [34.40385583372232]
  Inverse reinforcement learning can enable generalization to new states by learning a parameterized reward function. Existing safe imitation learning approaches based on IRL deal with this uncertainty using a maxmin framework. BROIL provides a natural way to interpolate between return-maximizing and risk-minimizing behaviors.
  arXiv  Detail & Related papers  (2020-07-24T01:52:11Z)
• Stochastic Linear Bandits Robust to Adversarial Attacks [117.665995707568]
  We provide two variants of a Robust Phased Elimination algorithm, one that knows $C$ and one that does not. We show that both variants attain near-optimal regret in the non-corrupted case $C = 0$, while incurring additional additive terms respectively. In a contextual setting, we show that a simple greedy algorithm is provably robust with a near-optimal additive regret term, despite performing no explicit exploration and not knowing $C$.
  arXiv  Detail & Related papers  (2020-07-07T09:00:57Z)
• Risk-Sensitive Reinforcement Learning: Near-Optimal Risk-Sample Tradeoff in Regret [115.85354306623368]
  We study risk-sensitive reinforcement learning in episodic Markov decision processes with unknown transition kernels. We propose two provably efficient model-free algorithms, Risk-Sensitive Value Iteration (RSVI) and Risk-Sensitive Q-learning (RSQ) We prove that RSVI attains an $tildeObig(lambda(|beta| H2) cdot sqrtH3 S2AT big)$ regret, while RSQ attains an $tildeObig(lambda
  arXiv  Detail & Related papers  (2020-06-22T19:28:26Z)

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.