Research map / Bandits and online learning / RL theory and MDPs
Variational Bayesian Reinforcement Learning with Regret Bounds
Brendan O'Donoghue
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Replace epsilon-greedy dithering with soft-Q-learning | Q-learning · DeepSea | max depth solved in 10^5 episodes: 6 -> 14 | Improved |
| Use K-learning (uncertainty-aware values) instead of soft Q-learning | soft Q-learning · DeepSea L=50 | time to solve | Improved |
| Optimized temperature tau* instead of scheduled temperature | K-learning · DeepSea | time to solve | Improved |
| Principled exploration (Thompson sampling, K-learning) instead of RLSVI, UCBVI, OQL, BEB | exploration strategy · DeepSea | time to solve | Improved |
| Remove optimism bonus and use a fixed temperature | K-learning · DeepSea | time to solve | Hurt |
Closest papers by meaning
- Risk-Sensitive Reinforcement Learning: Near-Optimal Risk-Sample Tradeoff in Regret (NeurIPS 2020 · similarity 0.90)
- Efficient Exploration in Average-Reward Constrained Reinforcement Learning: Achieving Near-Optimal Regret With Posterior Sampling (ICML 2024 · similarity 0.90)
- Optimistic posterior sampling for reinforcement learning: worst-case regret bounds (NeurIPS 2017 · similarity 0.90)
- Minimax Optimal Reinforcement Learning with Quasi-Optimism (ICLR 2025 · similarity 0.89)
- Bayesian Exploration Networks (ICML 2024 · similarity 0.89)
- Provably Efficient Partially Observable Risk-sensitive Reinforcement Learning with Hindsight Observation (ICML 2024 · similarity 0.88)
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.
