Research map / Reinforcement learning / Visual and representation RL
Where to Add Actions in Human-in-the-Loop Reinforcement Learning
Travis Mandel, Yun-En Liu, Emma Brunskill, Zoran Popović
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 |
|---|---|---|---|
| select states with ELI instead of uniformly random state (ablation) | PSRL with added actions · Riverswim | cumulative reward | Improved |
| select states with ELI instead of uniformly random state (ablation) | PSRL with added actions · Marblemaze | cumulative reward | Improved |
| select states with ELI instead of uniformly random state (ablation) | PSRL with added actions · Large Action Task | cumulative reward | Improved |
| select states by visit frequency instead of uniformly random (ablation) | PSRL with added actions · Riverswim | cumulative reward | Hurt |
| replace ELI-Limit cap with Beta-posterior probability of improvement (ablation) | ELI · Riverswim | cumulative reward | Improved |
| use UCRL confidence sets as optimistic estimator (ablation) | ELI · Riverswim | cumulative reward | Hurt |
| use MBIE optimistic estimator instead of BOSS (ablation) | ELI · Riverswim | cumulative reward | Mixed |
| use BOSS instead of Thompson sampling optimism (ablation) | ELI · Marblemaze with 96 uninformative outcomes | cumulative reward | Improved |
| use ELI under an improving simulated expert (ablation) | ELI · Large Action Task, improving expert | cumulative reward | Improved |
| use ELI with a poor simulated expert (ablation) | ELI · Large Action Task, poor expert | cumulative reward | Improved |
Closest papers by meaning
- GUIDE: Real-Time Human-Shaped Agents (NeurIPS 2024 · similarity 0.86)
- Leveraging Sub-Optimal Data for Human-in-the-Loop Reinforcement Learning (ICLR 2025 · similarity 0.86)
- Exploring Data Scaling Trends and Effects in Reinforcement Learning from Human Feedback (NeurIPS 2025 · similarity 0.85)
- Behavioral Exploration: Learning to Explore via In-Context Adaptation (ICML 2025 · similarity 0.84)
- Eureka: Human-Level Reward Design via Coding Large Language Models (ICLR 2024 · similarity 0.84)
- Efficient Imitation under Misspecification (ICLR 2025 · similarity 0.84)
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.
