Research map / Bandits and online learning / Contextual bandits and Thompson sampling

On Stochastic Contextual Bandits with Knapsacks in Small Budget Regime

ICLR 2025 · Bandits and online learning › Contextual bandits and Thompson sampling · approach: Oracle-efficient contextual bandits

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Studies stochastic contextual bandits with knapsacks (CBwK) in the small budget regime (B = Omega(sqrt(T))) and proposes algorithms for this setting.

Problem

Stochastic contextual bandits with knapsack constraints, where a learner must maximize cumulative reward over T rounds under a budget B, with focus on the small budget regime B = Omega(sqrt(T)).

Approach

Studies CBwK specifically in the small budget regime where the budget scales as Omega(sqrt(T)), rather than the large budget regime commonly assumed; the abstract is truncated so the specific algorithmic contribution is not stated.

Closest papers by meaning

• Linear Contextual Bandits with Knapsacks (NeurIPS 2016 · similarity 0.93)
• No-Regret is not enough! Bandits with General Constraints through Adaptive Regret Minimization (ICML 2025 · similarity 0.92)
• Beyond Primal-Dual Methods in Bandits with Stochastic and Adversarial Constraints (NeurIPS 2024 · similarity 0.91)
• High-dimensional Linear Bandits with Knapsacks (ICML 2024 · similarity 0.91)
• Contextual Decision-Making with Knapsacks Beyond the Worst Case (NeurIPS 2024 · similarity 0.90)
• Logistic Bandits with $\tilde{O}(\sqrt{dT})$ Regret without Context Diversity Assumptions (NeurIPS 2026 · similarity 0.90)