New algorithm reduces worst-case regret for heavy-tailed bandits.
problem Stochastic Multi-Armed Bandit problem with heavy-tailed rewards.
method Modified minimax policy MOSS with saturated empirical mean.
result Worst-case regret matching lower bound for heavy-tailed distributions.
New bandit problem for finding best group of arms with worst mean reward.
problem Finding the best group of arms with the worst mean reward in overlapping groups.
method Two algorithms based on successive elimination and robust optimization.
result Upper bounds on the number of samples to find max-min optimal or near-optimal group.
This paper analyzes how randomizing rewards in MBRL can improve performance without being overly optimistic.
problem The gap between theoretical worst-case regret analysis and empirical performance in MBRL.
method Reward randomization in model-based reinforcement learning (MBRL) with kernelized linear regulator (KNR) model.
result Reward randomization guarantees partial optimism and near-optimal worst-case regret.
This work analyzes the value of future reward information in RL.
problem Analyzing the impact of knowing future rewards in reinforcement learning.
method Competitive analysis and worst-case reward distribution.
result Exact ratios between standard RL agents and those with future-reward lookahead.
Identifies arms with rewards significantly different from the majority.
problem Identifying arms with rewards deviating substantially from the majority.
method Computing outlier threshold using median and median absolute deviation.
result Proposes two δ-PAC algorithms for ROAI with upper bounds on sample complexity and worst case lower bound.
Online learning has traditionally focused on the expected rewards. In this paper, a risk-averse online learning problem under the performance measure of the mean-variance of the rewards is studied. Both the bandit and full information settings are considered. The performance of several existing policies is analyzed, an…
The paper shows how policy regularization acts like an adversary to improve robustness.
problem Improving robustness of learned policies in reinforcement learning.
method Using convex duality, the paper characterizes adversarial reward perturbations and provides generalization guarantees.
result Policy regularization acts as an adversary to improve robustness against worst-case reward perturbations.
Paper tackles robustness in reward learning with partial identifiability.
problem Partial identifiability in reward learning leads to unreliable target reward recovery.
method Introduces a robust approach to maximize performance with respect to the worst-case reward in the feasible set.
result Develops Rob-ReL, an algorithm that maximizes performance under worst-case identifiability conditions.
The paper tackles a bandit problem with infinitely many arms per group, aiming to identify the group with the highest quantile reward.
problem Max-quantile group bandit problem with infinitely many arms per group.
method Two-step algorithm: first request arms from each group, then apply a finite-arm max-quantile bandit algorithm.
result Characterization of instance-dependent and worst-case regret, with matching lower bounds.
Flexible algorithms for maximizing rewards in structured bandits.
problem Reward maximization in structured stochastic multi-armed bandit problems.
method Asymptotically optimal algorithms using iterative saddle-point solvers.
result Achieves optimal performance with minimal computational burden.
Proposes DRRO to mitigate over-optimization in RLHF from human feedback.
problem Over-optimization due to reward misspecification in RLHF.
method Wasserstein distributionally robust regret optimization (DRRO).
result DRRO mitigates over-optimization more effectively than existing baselines.
Framework for robust control in cooperative systems with uncertain common noise.
problem Optimizing collective behavior of agents in the presence of uncertain common noise.
method Proposes a robust mean-field control framework and proves existence of optimal controls.
result Existence of optimal open-loop controls linked to a lifted robust Markov decision problem.
New algorithm reduces regret in infinitely many-armed bandits with decreasing rewards.
problem Infinitely many-armed bandits with rotting rewards.
method UCB index and adaptive threshold for unknown rotting rate, UCB index alone for known rotting rate.
result Matching upper bounds on regret achieved for different scenarios.
New algorithm improves online decision making by adaptively inferring arm rewards.
problem Adaptive selection of arms in MAB leads to non-iid data, complicating accurate inference.
method Proposes a doubly adaptive TS algorithm that leverages causal inference for adaptive reweighting.
result Demonstrates superior empirical performance in identifying the best arm compared to UCB and TS.
A new algorithm avoids worst-case outcomes in risky contexts.
problem Risk-averse behavior in contextual bandits is challenging.
method Developed a first risk-averse contextual bandit algorithm with online regret guarantees.
result First algorithm with an online regret guarantee for risk-averse contextual bandits.
New approach to multi-armed bandit problem aims to maximize highest total reward.
problem Traditional multi-armed bandit problem objective of maximizing total reward is not suitable in certain applications.
method Adaptive explore-then-commit policy with confidence bounds and adaptive stopping criterion.
result Achieves asymptotic and worst-case regret bounds for the new objective.
Algorithm allocates budgets to tasks with semi-bandit feedback, achieving near-optimal regret bounds.
problem Stochastic budget allocation with censored semi-bandit feedback.
method Optimism-based algorithm operating under censored semi-bandit feedback.
result Regret scales polylogarithmically with horizon T in diminishing-returns regimes.
A scheme robust to action erasures improves MAB performance.
problem Learning in multi-arm bandits with erasure channels.
method Proposes a scheme to work on top of any MAB algorithm.
result Worst-case regret is at most a factor of O ( 1 / 1 − ε ) O(1/\sqrt{1-ε}) O ( 1/ 1 − ε ) away from no-erasure case. Algorithm tackles adaptive discretization in adversarial Lipschitz bandits for dynamic pricing and auctions.
problem Adaptive discretization in adversarial Lipschitz bandits.
method Adversarial Zooming algorithm for adaptive discretization.
result First algorithm for adversarial Lipschitz bandits with instance-dependent regret bounds.
UCB algorithm's arm-sampling behavior is revealed, leading to new insights and proofs.
problem Optimizing multi-armed bandit algorithms for worst-case scenarios.
method Analysis of UCB algorithm's arm-sampling behavior and process-level characterization.
result UCB's arm-sampling rates are asymptotically deterministic, regardless of problem complexity.
Learning reward functions can lead to poor policy performance despite low error.
problem Low error in learned reward functions does not guarantee low regret in policy performance.
method Mathematical analysis of reward learning and policy optimization.
result A low expected test error of the reward model guarantees low worst-case regret, but error-regret mismatch can occur with certain data distributions.
KL-MS improves regret bounds for multi-armed bandits with bounded rewards.
problem Designing efficient exploration algorithms for multi-armed bandits with bounded rewards.
method Kullback-Leibler Maillard Sampling (KL-MS) for multi-armed bandits with bounded rewards.
result KL-MS achieves a worst-case regret bound of O ( μ ∗ ( 1 − μ ∗ ) K T ln K + K ln T ) O(\sqrt{μ^*(1-μ^*) K T \ln K} + K \ln T) O ( μ ∗ ( 1 − μ ∗ ) K T ln K + K ln T ) . The paper develops a robust algorithm for contextual bandits with heavy-tailed rewards.
problem Contextual bandits with heavy-tailed rewards.
method Develops an algorithm based on Catoni's estimator for robust statistics, applying it to contextual bandits with general function approximation.
result Establishes regret bounds that depend on cumulative reward variance and logarithmically on the reward range and number of rounds.
New approach limits regret in non-stationary bandits.
problem Understanding worst case regret in time-varying bandits.
method Belief inertia argument to resist new evidence after changes.
result Linear worst case regret for classical and restarting algorithms.
Regularized policies are robust to adversarial rewards.
problem Understanding the effects of regularization on policy exploration and robustness.
method Using Fenchel duality to derive the dual problem of the regularized RL objective, showing the optimal policy is robust to adversarial rewards.
result Regularized policies are optimal for a reinforcement learning problem under adversarial reward conditions.
New algorithm tackles heavy-tailed rewards in RL with instance-dependent regret bounds.
problem Efficient algorithms for RL with heavy-tailed rewards in large state-action spaces.
method Design of \textsc{Heavy-OFUL} for heavy-tailed linear bandits and \textsc{Heavy-LSVI-UCB} for RL with linear function approximation.
result First instance-dependent regret bounds for heavy-tailed rewards in RL with linear function approximation.
Paper introduces DOO models to outperform SAA out-of-sample.
problem Outperforming SAA in out-of-sample performance.
method Introduces DOO models that consider both worst-case and best-case scenarios.
result DOO models can always outperform SAA out-of-sample.
New models for bandit problems with fidelity rewards are introduced and analyzed.
problem Fidelity rewards in bandit problems to incentivize loyalty.
method Two models (loyalty-points and subscription) for fidelity rewards; stochastic and adversarial settings considered.
result Sublinear regret bounds for some models, worst case lower bounds for others.
Study non-linear combinatorial bandits with polynomial rewards, finding significant differences from linear cases.
problem Adversarial combinatorial bandits with general non-linear reward functions.
method Extending existing work on adversarial linear combinatorial bandits, analyzing minimax optimal regret for polynomial and non-polynomial reward functions.
result Minimax optimal regret bounds for adversarial combinatorial bandits with general non-linear reward functions.
In this paper we consider the worst-case model risk approach described in Glasserman and Xu (2014). Portfolio selection with model risk can be a challenging operational research problem. In particular, it presents an additional optimisation compared to the classical one. We find the analytical solution for the optimal …
This paper calculates worst-case target semi-variances for uncertain losses.
problem Managing risk when loss distribution is uncertain and only partial information is known.
method Derives worst-case target semi-variances for symmetric or non-negative losses under uncertainty sets representing investor's undesirable scenarios.
result Closed-form expressions for worst-case target semi-variances are derived.
Robust hypothesis testing designs a test for worst-case distributions using kernel methods.
problem Design a robust test for hypothesis testing under uncertainty sets.
method Data-driven uncertainty sets constructed using kernel mean embeddings and maximum mean discrepancy (MMD). Bayesian and Neyman-Pearson settings investigated.
result Proposed robust kernel tests are exponentially consistent and asymptotically optimal.
The focal point of this paper is the issue of "drawdown" which arises in recursive betting scenarios and related applications in the stock market. Roughly speaking, drawdown is understood to mean drops in wealth over time from peaks to subsequent lows. Motivated by the fact that this issue is of paramount concern to co…
IDS optimizes regret in stochastic partial monitoring with linear rewards.
problem Optimizing decision-making in uncertain environments with linear rewards.
method Information Directed Sampling (IDS) for stochastic partial monitoring.
result Achieves optimal regret rates in all observable game regimes.
Worst-Case Sensitivity measures model sensitivity to uncertainty set size.
problem Model sensitivity to uncertainty set size in Distributionally Robust Optimization.
method Introducing Worst-Case Sensitivity as a measure of model sensitivity, and deriving closed-form expressions for various uncertainty sets.
result DRO solutions can be sensitive to the family and size of the uncertainty set, and worst-case sensitivity reflects these properties.
This paper formalizes reward training in RLHF and provides theoretical guarantees.
problem Costly human feedback collection in RLHF.
method Linear contextual dueling bandit method for regret minimization.
result Derives bounds on simple regret for offline reward training.
A new method resolves non-identifiability in reward modeling using anchor labels.
problem Non-identifiability in reward modeling from pairwise preferences alone.
method Anchor-guided Variance-aware Reward Modeling (AVRM) framework.
result AVRM resolves non-identifiability and improves reward modeling performance.
Study non-rectangular robust MDPs for average-reward, finding optimal policies and transient values.
problem Non-rectangular robust Markov decision processes under average-reward criterion.
method Proves history-dependent policies are robust-optimal, introduces transient-value framework, constructs epoch-based policy.
result Existence and properties of robust optimal policies, transient value bounds.
RADIAL-RL improves deep RL agents' robustness against adversarial attacks.
problem Vulnerability of deep reinforcement learning agents to small adversarial perturbations.
method RADIAL-RL, a principled framework for training robust reinforcement learning agents.
result RADIAL-RL-trained agents consistently outperform prior methods in robustness tests.
New method for identifying best designs in vector optimization with uncertain feedback.
problem Optimizing vector-valued outcomes with uncertain preferences.
method Stochastic bandit feedback, polyhedral ordering cone, ( ε , δ ε,δ ε , δ )-PAC Pareto set identification. result Sample complexity characterized and matched by the naïve elimination algorithm.
Improved algorithms for stochastic linear bandits using tighter confidence sequences.
problem Stochastic linear bandits with improved worst-case regret guarantees.
method Novel tail bound for adaptive martingale mixtures to construct tighter confidence sequences.
result Linear bandit algorithm achieves competitive worst-case regret.
Model shows how centralization occurs in cryptocurrency mining.
problem Centralization of reward and computational power in Bitcoin-like cryptocurrencies.
method Mean field game model to study miner competition and reward distribution.
result Heterogeneity of initial wealth leads to greater imbalance in reward distribution.
We propose a bandit algorithm that explores by randomizing its history of rewards. Specifically, it pulls the arm with the highest mean reward in a non-parametric bootstrap sample of its history with pseudo rewards. We design the pseudo rewards such that the bootstrap mean is optimistic with a sufficiently high probabi…
GRPO optimizes LLMs with verifiable rewards, amplifying policy success.
problem Improving LLMs' reasoning under verifiable binary rewards.
method Introduces GRPO, analyzes variants of reward normalization and regularization.
result GRPO amplifies policy success, converging to a fixed point exceeding the reference.
The paper analyzes worst-case distortion risk metrics and weighted entropy under partial information.
problem Analyzing worst-case distortion risk metrics and weighted entropy with limited information.
method General distributions, partial information (mean and variance), various entropies and risk measures.
result Provides worst-case results for distortion risk metrics and weighted entropy.
AceIRL learns expert reward from active exploration.
problem Learning reward function from expert demonstrations in unknown environments.
method Active exploration to infer reward function and identify good policy.
result First approach to active IRL with sample-complexity bounds that doesn't require a generative model.
Paper tackles robust offline RL with heavy-tailed rewards.
problem Real-world applications often encounter heavy-tailed rewards, challenging offline RL.
method Proposes ROAM and ROOM algorithms using median-of-means method for robust off-policy evaluation and OPO.
result Demonstrates superior performance on heavy-tailed reward datasets compared to existing methods.
We study a variant of the stochastic K K K -armed bandit problem, which we call "bandits with delayed, aggregated anonymous feedback". In this problem, when the player pulls an arm, a reward is generated, however it is not immediately observed. Instead, at the end of each round the player observes only the sum of a number…