Research map / Bandits and online learning / Online convex optimization and games
Thompson Sampling for Stochastic Bandits with Graph Feedback
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 TS-MaxN instead of UCB-N or UCB-MaxN | Stochastic bandits with Erdős–Rényi graph feedback · Erdős–Rényi graphs, sparse and dense | cumulative regret | Helped |
| Use TS-N instead of UCB-N or UCB-MaxN | Stochastic bandits with Erdős–Rényi graph feedback · Erdős–Rényi graphs, sparse and dense | cumulative regret | Helped |
| Choose the empirically best arm in the sampled arm's neighborhood (ablation) | TS-N · Erdős–Rényi graphs, sparse and dense | cumulative regret | Helped |
| Use a dense rather than sparse feedback graph | All evaluated policies · Erdős–Rényi graphs | cumulative regret | Helped |
| Use TS-MaxN instead of the other evaluated policies | Stochastic bandits with power-law graph feedback · Power-law graphs, dense and sparse | cumulative regret | Helped |
| Use TS-N instead of UCB-MaxN | Stochastic bandits with sparse power-law graph feedback · Power-law graphs, sparse | cumulative regret | Mixed |
| Increase the number of planted groups | Evaluated bandit policies · Planted partition graphs | relative regret | Mixed |
| Use Thompson sampling instead of the evaluated alternatives | Stochastic bandits with inferred Facebook graph feedback · Facebook | cumulative regret | Helped |
| Use Thompson sampling instead of UCB-N or UCB-MaxN | Stochastic bandits with inferred Flixster graph feedback · Flixster | cumulative regret | Helped |
| Optimize EPSGREEDYD's exploration hyperparameters post hoc (ablation) | EPSGREEDYD · Flixster | cumulative regret | Helped |
Closest papers by meaning
- Information Directed Sampling for Stochastic Bandits With Graph Feedback (AAAI 2018 · similarity 0.95)
- Stochastic Bandits with Graph Feedback in Non-Stationary Environments (AAAI 2021 · similarity 0.92)
- Stochastic contextual bandits with graph feedback: from independence number to MAS number (NeurIPS 2024 · similarity 0.91)
- Thompson Sampling For Combinatorial Bandits: Polynomial Regret and Mismatched Sampling Paradox (NeurIPS 2024 · similarity 0.90)
- Feel-Good Thompson Sampling for Contextual Bandits: a Markov Chain Monte Carlo Showdown (NeurIPS 2025 · similarity 0.89)
- Adversarial Combinatorial Semi-bandits with Graph Feedback (ICML 2025 · similarity 0.89)
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.