Research map / Reinforcement learning / Visual and representation RL
Curiosity in Hindsight: Intrinsic Exploration in Stochastic Environments
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 |
|---|---|---|---|
| Add hindsight representations, reconstruction, and an invariance objective to BYOL-Explore | BYOL-Explore · Pycolab Maze, Brownian Oscillators | trackers reached | Helped |
| Add hindsight representations, reconstruction, and an invariance objective to BYOL-Explore | BYOL-Explore · Pycolab Maze, Random Pixel Noise | trackers reached | Helped |
| Add hindsight representations, reconstruction, and an invariance objective to BYOL-Explore | BYOL-Explore · Pycolab Maze, On-Demand Pixel Noise | trackers reached | Helped |
| Add hindsight representations, reconstruction, and an invariance objective to BYOL-Explore | BYOL-Explore · Pycolab Maze, noise-free Baseline | trackers reached | No clear effect |
| Add hindsight representations, reconstruction, and an invariance objective to BYOL-Explore | BYOL-Explore · Bank Heist, Natural Traps and Sticky Actions | extrinsic reward per episode | Helped |
| Add hindsight representations, reconstruction, and an invariance objective to BYOL-Explore | BYOL-Explore · Montezuma's Revenge, Sticky Actions | rooms discovered | Helped |
| Add hindsight representations, reconstruction, and an invariance objective to BYOL-Explore | BYOL-Explore · Montezuma's Revenge, Non-Sticky Baseline | rooms discovered | No clear effect |
| Add hindsight representations, reconstruction, and an invariance objective to BYOL-Explore | BYOL-Explore · Hard-exploration Atari games, Sticky Actions | extrinsic reward | Helped |
| Add hindsight representations, reconstruction, and an invariance objective to BYOL-Explore | BYOL-Explore · Persistive Noise | episode return | Helped |
| Use a lower contrastive-loss temperature rather than a higher temperature (ablation) | BYOL-Hindsight · Montezuma's Revenge, Non-Sticky Baseline, intrinsic-only | exploration performance | Helped |
| Vary the contrastive-loss temperature (ablation) | BYOL-Hindsight · Sticky Actions | exploration performance | No clear effect |
Closest papers by meaning
- Large-Scale Study of Curiosity-Driven Learning (ICLR 2019 · similarity 0.91)
- Wonder Wins Ways: Curiosity-Driven Exploration through Multi-Agent Contextual Calibration (NeurIPS 2025 · similarity 0.89)
- Decoupling Exploration and Policy Optimization: Uncertainty Guided Tree Search for Hard Exploration (NeurIPS 2026 · similarity 0.87)
- Action-Dependent Optimality-Preserving Reward Shaping (ICML 2025 · similarity 0.87)
- Beyond Optimism: Exploration With Partially Observable Rewards (NeurIPS 2024 · similarity 0.87)
- DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning (NeurIPS 2025 · 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.