Research map / Reinforcement learning / Visual and representation RL
PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training
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 PEBBLE instead of Preference PPO with a budget of 1400 queries | Hopper · Gym Hopper | performance relative to the respective oracle: 1400 queries | Helped |
| add pre-training (ablation) | PrePPO · Quadruped | performance | Helped |
| add pre-training (ablation) | PrePPO · Walker | performance | Helped |
| use PEBBLE rather than SAC with ground-truth reward | Walker · Walker, 1M timesteps | wall-clock time: 5 hours 26 mins -> 5 hours 47 minutes | Hurt |
| use PEBBLE for human feedback | Hopper backflip · Hopper backflip | number of queries: less than half as many queries | Helped |
Closest papers by meaning
- Leveraging Sub-Optimal Data for Human-in-the-Loop Reinforcement Learning (ICLR 2025 · similarity 0.92)
- DUO: Diverse, Uncertain, On-Policy Query Generation and Selection for Reinforcement Learning from Human Feedback (AAAI 2025 · similarity 0.91)
- GUIDE: Real-Time Human-Shaped Agents (NeurIPS 2024 · similarity 0.90)
- Enhancing Rating-Based Reinforcement Learning to Effectively Leverage Feedback from Large Vision-Language Models (ICML 2025 · similarity 0.89)
- Exploring Data Scaling Trends and Effects in Reinforcement Learning from Human Feedback (NeurIPS 2025 · similarity 0.89)
- A Theoretical Framework for Partially-Observed Reward States in RLHF (ICLR 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.