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

2026-02-27

AI Summary

This work addresses the problem of best-arm identification under resource constraints, where each pull of an arm consumes heterogeneous and limited resources. To tackle this challenge, the authors propose the SH-RR algorithm, which integrates a resource-aware rationing mechanism into the successive halving framework for the first time. The approach unifies the treatment of both stochastic and deterministic resource consumption scenarios and introduces a novel effective resource consumption metric. Theoretical analysis demonstrates that SH-RR efficiently identifies the optimal arm across diverse resource consumption patterns, providing a unified sample complexity guarantee and significantly improving identification efficiency in resource-constrained settings.

Technology Category

• Reasoning under Uncertainty: Stochastic Optimization
• Machine Learning: Online Learning & Bandits
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

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

Problem

• Best Arm Identification
• Resource Constraints
• Budgeted Learning
• Multi-armed Bandits

Innovation

• Best Arm Identification
• Resource Constraints
• Successive Halving
• Resource Allocation
• Effective Consumption Measure