arXiv subjects

Wang Chi Cheung

Publications and source records attributed to Wang Chi Cheung.

At least 19 recordsLinked to original sources

arXiv2026

Pure Exploration for a Good Policy in Reinforcement Learning with Bandit Feedback

Pure exploration in episodic Reinforcement Learning has primarily focused on Best Policy Identification (BPI), which seeks to identify a (near)-optimal policy with high confidence. Motivated by practical settings where a ``good enough'' policy suffices, we study an alternate objective of Good Policy Identification (GPI). For a given reward threshold $μ_0$, GPI only requires identifying a policy with expected reward in an episode at least $μ_0$ if such a policy exists (positive instance), or declaring None if no such policy exists (negative instance). We formalize GPI under the fixed-confidence setting. We require the output to be correct with probability $\geq 1-δ$, and seek to minimize the expected sample complexity, which is the expected number of episodes explored for the output. We propose a novel algorithm BEE-GPI, and derive theoretically-grounded upper bounds on its sample complexity for positive and negative instances. Notably, for positive instances, the coefficient of $\log 1/δ$ in our upper bound is $O(H^2/(V^* - μ_0)^2)$, where $H$ is the episode length and $V^*$ is the optimal expected reward in an episode. The coefficient does not depend on the action and state space sizes otherwise, in sharp contrast to the sample complexity in BPI. We further establish lower bound results to show the near-optimality of BEE-GPI and the necessity of the $1/(V^* -μ)^2$ term. Numerical experiments further validate the efficiency of our approach.

cs.LG↗

arXiv2026

Closing the Gap on the Sample Complexity of 1-Identification

The 1-identification problem is a fundamental pure-exploration problem in multi-armed bandits. An agent aims to determine whether there exists an arm whose mean reward exceeds a known threshold $μ_0$, or to output \textsf{None} otherwise. The agent must guarantee correctness with probability at least $1-δ$, while minimizing the expected number of arm pulls $\mathbb{E}[τ]$. We study the 1-identification problem and make two main contributions. First, for instances with at least one qualified arm, we derive a new lower bound on $\mathbb{E}[τ]$ via a novel optimization formulation. Second, we propose a new algorithm and establish upper bounds that match the lower bounds up to polynomial logarithmic factors uniformly over all instances. Our result complements the analysis of $\mathbb{E}τ$ when there are multiple qualified arms, which is an open problem in the literature.

cs.LG↗

arXiv2026

Learning with a Budget: Identifying the Best Arm with Resource Constraints

In many applications, evaluating the effectiveness of different alternatives comes with varying costs or resource usage. Motivated by such heterogeneity, we study the Best Arm Identification with Resource Constraints (BAIwRC) problem, where an agent seeks to identify the best alternative (aka arm) in the presence of resource constraints. Each arm pull consumes one or more types of limited resources. We make two key contributions. First, we propose the Successive Halving with Resource Rationing (SH-RR) algorithm, which integrates resource-aware allocation into the classical successive halving framework on best arm identification. The SH-RR algorithm unifies the theoretical analysis for both the stochastic and deterministic consumption settings, with a new \textit{effective consumption measure

cs.LG↗

arXiv2026

Episodic Contextual Bandits with Knapsacks under Conversion Models

We study an online setting, where a decision maker (DM) interacts with contextual bandit-with-knapsack (BwK) instances in repeated episodes. These episodes start with different resource amounts, and the contexts' probability distributions are non-stationary in an episode. All episodes share the same latent conversion model, which governs the random outcome contingent upon a request's context and an allocation decision. Our model captures applications such as dynamic pricing on perishable resources with episodic replenishment, and first price auctions in repeated episodes with different starting budgets. We design an online algorithm that achieves a regret sub-linear in $T$, the number of episodes, assuming access to a \emph{confidence bound oracle} that achieves an $o(T)$-regret. Such an oracle is readily available from existing contextual bandit literature. We overcome the technical challenge with arbitrarily many possible contexts, which leads to a reinforcement learning problem with an unbounded state space. Our framework provides improved regret bounds in certain settings when the DM is provided with unlabeled feature data, which is novel to the contextual BwK literature.

cs.LG↗

arXiv2025

Online Bandits with (Biased) Offline Data: Adaptive Learning under Distribution Mismatch

Traditional online learning models are typically initialized from scratch. By contrast, contemporary real-world applications often have access to historical datasets that can potentially enhanced the online learning processes. We study how offline data can be leveraged to facilitate online learning in stochastic multi-armed bandits and combinatorial bandits. In our study, the probability distributions that govern the offline data and the online rewards can be different. We first show that, without a non-trivial upper bound on their difference, no non-anticipatory policy can outperform the classical Upper Confidence Bound (UCB) policy, even with the access to offline data. In complement, we propose an online policy MIN-UCB for multi-armed bandits. MIN-UCB outperforms the UCB when such an upper bound is available. MIN-UCB adaptively chooses to utilize the offline data when they are deemed informative, and to ignore them otherwise. We establish that MIN-UCB achieves tight regret bounds, in both instance independent and dependent settings. We generalize our approach to the combinatorial bandit setting by introducing MIN-COMB-UCB, and we provide corresponding instance dependent and instance independent regret bounds. We illustrate how various factors, such as the biases and the size of offline datasets, affect the utility of offline data in online learning. We discuss several applications and conduct numerical experiments to validate our findings.

cs.LG↗

arXiv2025

Near Optimal Non-asymptotic Sample Complexity of 1-Identification

Motivated by an open direction in existing literature, we study the 1-identification problem, a fundamental multi-armed bandit formulation on pure exploration. The goal is to determine whether there exists an arm whose mean reward is at least a known threshold $μ_0$, or to output None if it believes such an arm does not exist. The agent needs to guarantee its output is correct with probability at least $1-δ$. Degenne & Koolen 2019 has established the asymptotically tight sample complexity for the 1-identification problem, but they commented that the non-asymptotic analysis remains unclear. We design a new algorithm Sequential-Exploration-Exploitation (SEE), and conduct theoretical analysis from the non-asymptotic perspective. Novel to the literature, we achieve near optimality, in the sense of matching upper and lower bounds on the pulling complexity. The gap between the upper and lower bounds is up to a polynomial logarithmic factor. The numerical result also indicates the effectiveness of our algorithm, compared to existing benchmarks.

cs.LG↗

arXiv2025

Best Arm Identification with Possibly Biased Offline Data

We study the best arm identification (BAI) problem with potentially biased offline data in the fixed confidence setting, which commonly arises in real-world scenarios such as clinical trials. We prove an impossibility result for adaptive algorithms without prior knowledge of the bias bound between online and offline distributions. To address this, we propose the LUCB-H algorithm, which introduces adaptive confidence bounds by incorporating an auxiliary bias correction to balance offline and online data within the LUCB framework. Theoretical analysis shows that LUCB-H matches the sample complexity of standard LUCB when offline data is misleading and significantly outperforms it when offline data is helpful. We also derive an instance-dependent lower bound that matches the upper bound of LUCB-H in certain scenarios. Numerical experiments further demonstrate the robustness and adaptability of LUCB-H in effectively incorporating offline data.

cs.LG↗

arXiv2025

Best Arm Identification with Resource Constraints

Motivated by the cost heterogeneity in experimentation across different alternatives, we study the Best Arm Identification with Resource Constraints (BAIwRC) problem. The agent aims to identify the best arm under resource constraints, where resources are consumed for each arm pull. We make two novel contributions. We design and analyze the Successive Halving with Resource Rationing algorithm (SH-RR). The SH-RR achieves a near-optimal non-asymptotic rate of convergence in terms of the probability of successively identifying an optimal arm. Interestingly, we identify a difference in convergence rates between the cases of deterministic and stochastic resource consumption.

cs.LG↗

arXiv2025

Efficiently Solving Discounted MDPs with Predictions on Transition Matrices

We study infinite-horizon Discounted Markov Decision Processes (DMDPs) under a generative model. Motivated by the Algorithm with Advice framework Mitzenmacher and Vassilvitskii 2022, we propose a novel framework to investigate how a prediction on the transition matrix can enhance the sample efficiency in solving DMDPs and improve sample complexity bounds. We focus on the DMDPs with $N$ state-action pairs and discounted factor $γ$. Firstly, we provide an impossibility result that, without prior knowledge of the prediction accuracy, no sampling policy can compute an $ε$-optimal policy with a sample complexity bound better than $\tilde{O}((1-γ)^{-3} Nε^{-2})$, which matches the state-of-the-art minimax sample complexity bound with no prediction. In complement, we propose an algorithm based on minimax optimization techniques that leverages the prediction on the transition matrix. Our algorithm achieves a sample complexity bound depending on the prediction error, and the bound is uniformly better than $\tilde{O}((1-γ)^{-4} N ε^{-2})$, the previous best result derived from convex optimization methods. These theoretical findings are further supported by our numerical experiments.

cs.LG↗

arXiv2023

Online Stochastic Allocation of Reusable Resources

We study a multi-objective model on the allocation of reusable resources under model uncertainty. Heterogeneous customers arrive sequentially according to a latent stochastic process, request for certain amounts of resources, and occupy them for random durations of time. The decision maker's goal is to simultaneously maximize multiple types of rewards generated by the customers, while satisfying the resource capacity constraints in each time step. We develop models and algorithms for deciding on the allocation actions. We show that when the usage duration is relatively small compared with the length of the planning horizon, our policy achieves $1-O(ε)$ fraction of the optimal expected rewards, where $ε$ decays to zero at a near optimal rate as the resource capacities grow.

math.OC↗

arXiv2023

Online Resource Allocation: Bandits feedback and Advice on Time-varying Demands

We consider a general online resource allocation model with bandit feedback and time-varying demands. While online resource allocation has been well studied in the literature, most existing works make the strong assumption that the demand arrival process is stationary. In practical applications, such as online advertisement and revenue management, however, this process may be exogenous and non-stationary, like the constantly changing internet traffic. Motivated by the recent Online Algorithms with Advice framework [Mitazenmacher and Vassilvitskii, \emph{Commun. ACM} 2022], we explore how online advice can inform policy design. We establish an impossibility result that any algorithm perform poorly in terms of regret without any advice in our setting. In contrast, we design an robust online algorithm that leverages the online predictions on the total demand volumes. Empowered with online advice, our proposed algorithm is shown to have both theoretical performance and promising numerical results compared with other algorithms in literature. We also provide two explicit examples for the time-varying demand scenarios and derive corresponding theoretical performance guarantees. Finally, we adapt our model to a network revenue management problem, and numerically demonstrate that our algorithm can still performs competitively compared to existing baselines.

cs.LG↗

arXiv2023

Achieving the Pareto Frontier of Regret Minimization and Best Arm Identification in Multi-Armed Bandits

We study the Pareto frontier of two archetypal objectives in multi-armed bandits, namely, regret minimization (RM) and best arm identification (BAI) with a fixed horizon. It is folklore that the balance between exploitation and exploration is crucial for both RM and BAI, but exploration is more critical in achieving the optimal performance for the latter objective. To this end, we design and analyze the BoBW-lil'UCB$(γ)$ algorithm. Complementarily, by establishing lower bounds on the regret achievable by any algorithm with a given BAI failure probability, we show that (i) no algorithm can simultaneously perform optimally for both the RM and BAI objectives, and (ii) BoBW-lil'UCB$(γ)$ achieves order-wise optimal performance for RM or BAI under different values of $γ$. Our work elucidates the trade-off more precisely by showing how the constants in previous works depend on certain hardness parameters. Finally, we show that BoBW-lil'UCB outperforms a close competitor UCB$_α$ (Degenne et al., 2019) in terms of the time complexity and the regret on diverse datasets such as MovieLens and Published Kinase Inhibitor Set.

cs.LG↗

arXiv2022

Online Resource Allocation for Reusable Resources

We study a general model on reusable resource allocation under model uncertainty. A heterogeneous population of customers arrive at the decision maker's (DM's) platform sequentially. Upon observing a customer's type, the DM selects an allocation decision, which leads to rewards earned and resources occupied. Each resource unit is occupied for a random duration, and the unit is available for another allocation after the usage duration. Our model captures numerous applications involving admission control and assortment planning. The DM aims to simultaneously maximize multiple types of rewards, while satisfying the resource constraints and being uncertain about the customers' arrival process. We develop a near-optimal algorithm that achieves $(1-ε)$ fraction of the optimal expected rewards, where the error parameter $ε$ decays to zero as the resource capacity units and the length of the horizon grow. The algorithm iteratively applies the Multiplicative Weight Update algorithm in a novel manner, which balances the trade-off among the amounts of rewards earned, resources occupied and usage durations.

math.OC↗

arXiv2021

Inventory Balancing with Online Learning

We study a general problem of allocating limited resources to heterogeneous customers over time under model uncertainty. Each type of customer can be serviced using different actions, each of which stochastically consumes some combination of resources, and returns different rewards for the resources consumed. We consider a general model where the resource consumption distribution associated with each (customer type, action)-combination is not known, but is consistent and can be learned over time. In addition, the sequence of customer types to arrive over time is arbitrary and completely unknown. We overcome both the challenges of model uncertainty and customer heterogeneity by judiciously synthesizing two algorithmic frameworks from the literature: inventory balancing, which "reserves" a portion of each resource for high-reward customer types which could later arrive, and online learning, which shows how to "explore" the resource consumption distributions of each customer type under different actions. We define an auxiliary problem, which allows for existing competitive ratio and regret bounds to be seamlessly integrated. Furthermore, we show that the performance guarantee generated by our framework is tight, that is, we provide an information-theoretic lower bound which shows that both the loss from competitive ratio and the loss for regret are relevant in the combined problem. Finally, we demonstrate the efficacy of our algorithms on a publicly available hotel data set. Our framework is highly practical in that it requires no historical data (no fitted customer choice models, nor forecasting of customer arrival patterns) and can be used to initialize allocation strategies in fast-changing environments.

cs.AI↗

arXiv2021

Learning to Optimize under Non-Stationarity

We introduce algorithms that achieve state-of-the-art \emph{dynamic regret} bounds for non-stationary linear stochastic bandit setting. It captures natural applications such as dynamic pricing and ads allocation in a changing environment. We show how the difficulty posed by the non-stationarity can be overcome by a novel marriage between stochastic and adversarial bandits learning algorithms. Defining $d,B_T,$ and $T$ as the problem dimension, the \emph{variation budget}, and the total time horizon, respectively, our main contributions are the tuned Sliding Window UCB (\texttt{SW-UCB}) algorithm with optimal $\widetilde{O}(d^{2/3}(B_T+1)^{1/3}T^{2/3})$ dynamic regret, and the tuning free bandit-over-bandit (\texttt{BOB}) framework built on top of the \texttt{SW-UCB} algorithm with best $\widetilde{O}(d^{2/3}(B_T+1)^{1/4}T^{3/4})$ dynamic regret.

cs.LG↗

arXiv2021

Probabilistic Sequential Shrinking: A Best Arm Identification Algorithm for Stochastic Bandits with Corruptions

We consider a best arm identification (BAI) problem for stochastic bandits with adversarial corruptions in the fixed-budget setting of T steps. We design a novel randomized algorithm, Probabilistic Sequential Shrinking($u$) (PSS($u$)), which is agnostic to the amount of corruptions. When the amount of corruptions per step (CPS) is below a threshold, PSS($u$) identifies the best arm or item with probability tending to $1$ as $T\rightarrow \infty$. Otherwise, the optimality gap of the identified item degrades gracefully with the CPS.We argue that such a bifurcation is necessary. In PSS($u$), the parameter $u$ serves to balance between the optimality gap and success probability. The injection of randomization is shown to be essential to mitigate the impact of corruptions. To demonstrate this, we design two attack strategies that are applicable to any algorithm. We apply one of them to a deterministic analogue of PSS($u$) known as Successive Halving (SH) by Karnin et al. (2013). The attack strategy results in a high failure probability for SH, but PSS($u$) remains robust. In the absence of corruptions, PSS($2$)'s performance guarantee matches SH's. We show that when the CPS is sufficiently large, no algorithm can achieve a BAI probability tending to $1$ as $T\rightarrow \infty$. Numerical experiments corroborate our theoretical findings.

cs.LG↗

arXiv2021

Thompson Sampling Algorithms for Cascading Bandits

Motivated by the pressing need for efficient optimization in online recommender systems, we revisit the cascading bandit model proposed by Kveton et al. (2015). While Thompson sampling (TS) algorithms have been shown to be empirically superior to Upper Confidence Bound (UCB) algorithms for cascading bandits, theoretical guarantees are only known for the latter. In this paper, we first provide a problem-dependent upper bound on the regret of a TS algorithm with Beta-Bernoulli updates; this upper bound is tighter than a recent derivation under a more general setting by Huyuk and Tekin (2019). Next, we design and analyze another TS algorithm with Gaussian updates, TS-Cascade. TS-Cascade achieves the state-of-the-art regret bound for cascading bandits. Complementarily, we consider a linear generalization of the cascading bandit model, which allows efficient learning in large cascading bandit problem instances. We introduce and analyze a TS algorithm, which enjoys a regret bound that depends on the dimension of the linear model but not the number of items. Finally, by using information-theoretic techniques and judiciously constructing cascading bandit instances, we derive a nearly matching regret lower bound for the standard model. Our paper establishes the first theoretical guarantees on TS algorithms for stochastic combinatorial bandit problem model with partial feedback. Numerical experiments demonstrate the superiority of the proposed TS algorithms compared to existing UCB-based ones.

cs.LG↗

arXiv2021

Hedging the Drift: Learning to Optimize under Non-Stationarity

We introduce data-driven decision-making algorithms that achieve state-of-the-art \emph{dynamic regret} bounds for non-stationary bandit settings. These settings capture applications such as advertisement allocation, dynamic pricing, and traffic network routing in changing environments. We show how the difficulty posed by the (unknown \emph{a priori} and possibly adversarial) non-stationarity can be overcome by an unconventional marriage between stochastic and adversarial bandit learning algorithms. Our main contribution is a general algorithmic recipe for a wide variety of non-stationary bandit problems. Specifically, we design and analyze the sliding window-upper confidence bound algorithm that achieves the optimal dynamic regret bound for each of the settings when we know the respective underlying \emph{variation budget}, which quantifies the total amount of temporal variation of the latent environments. Boosted by the novel bandit-over-bandit framework that adapts to the latent changes, we can further enjoy the (nearly) optimal dynamic regret bounds in a (surprisingly) parameter-free manner. In addition to the classical exploration-exploitation trade-off, our algorithms leverage the power of the "forgetting principle" in the learning processes, which is vital in changing environments. Our extensive numerical experiments on both synthetic and real world online auto-loan datasets show that our proposed algorithms achieve superior empirical performance compared to existing algorithms.

cs.LG↗