Research map / Bandits and online learning / Multi-armed bandits
Heterogeneous Multi-player Multi-armed Bandits: Closing the Gap and Generalization
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use BEACON instead of METC | Heterogeneous MP-MAB with linear reward · 5 arms, 5 players; reported horizon | regret: approximate 7 × regret reduction | Helped |
| Use BEACON instead of METC | Heterogeneous MP-MAB with linear reward · 100 randomly generated instances; M = 5, K =5, T = 10^6 | average regret: approximately 6× lower | Helped |
| Use BEACON instead of METC | Heterogeneous MP-MAB with linear reward · Heterogeneous MP-MAB | problem-dependent regret bound: O\left(\frac{M^3K}{\Delta_{\min}}\log(T)\right) -> \tilde{O}\left(\frac{M^2K}{\Delta_{\min}}\log(T)\right) | Helped |
| Use BEACON instead of the prior best known bound | Heterogeneous MP-MAB with linear reward · Heterogeneous MP-MAB | problem-independent regret bound: O\left(M^{\frac{3}{2}}\sqrt{KT\log(T)}\right) -> O\left(M\sqrt{KT\log(T)}\right) | Helped |
| Apply adaptive differential communication (ADC) | BEACON implicit communication · Per-epoch communication loss | communication loss: O(\log(T)) -> O(1) | Helped |
| Use CUCB-style UCB matching selection rather than matching elimination | BEACON exploration · Heterogeneous MP-MAB exploration | exploration loss | Helped |
| Use BEACON instead of METC | Heterogeneous MP-MAB with proportional-fairness and minimal rewards · M = 6, K = 8 | regret | Helped |
Closest papers by meaning
- Optimal Algorithm for Max-Min Fair Bandit (ICML 2025 · similarity 0.90)
- Observe Before Play: Multi-Armed Bandit with Pre-Observations (AAAI 2020 · similarity 0.90)
- Decentralized Scheduling with QoS Constraints: Achieving O(1) QoS Regret of Multi-Player Bandits (AAAI 2024 · similarity 0.89)
- Heterogeneous Multi-Agent Bandits with Parsimonious Hints (AAAI 2025 · similarity 0.88)
- Individual Regret in Cooperative Stochastic Multi-Armed Bandits (NeurIPS 2025 · similarity 0.88)
- Thompson Sampling for Real-Valued Combinatorial Pure Exploration of Multi-Armed Bandit (AAAI 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.