Research map / Reinforcement learning / Visual and representation RL
Modeling Human Exploration Through Resource-Rational Reinforcement Learning
Marcel Binz, Eric Schulz
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 |
|---|---|---|---|
| Use RR-RL2 instead of the hybrid probit exploration model to fit human choices | RR-RL2 · Two-armed bandit human data | BIC: 6158.91 -> 5562.63 | Improved |
| Increase description length of RR-RL2 from 1 to about 100 nats | RR-RL2 · Two-armed bandit probit analysis | exploration style: average regression coefficient of 0.15 | Mixed |
| Increase description length from 100 to 1000 nats | RR-RL2 · Two-armed bandit probit analysis | Thompson sampling weight: average regression coefficient of 0.19 | Improved |
| Increase description length to 1000 nats and above | RR-RL2 · Two-armed bandit probit analysis | UCB-based weight: average regression coefficient of 0.23 | Mixed |
| Reduce description length of RR-RL2 | RR-RL2 · Iowa Gambling Task | high-risk arm selection | Hurt |
| Increase description length of RR-RL2 | RR-RL2 · Horizon task | strategic directed exploration | Improved |
| Increase description length of RR-RL2 | RR-RL2 · Horizon task | strategic random exploration | No clear effect |
| Increase reward noise standard deviation by 50% when retraining (ablation) | RR-RL2 · Horizon task | interaction effect on directed exploration | Improved |
| Increase description length of RR-RL2 (regret) | RR-RL2 · Horizon task | regret | Improved |
Closest papers by meaning
- First-Explore, then Exploit: Meta-Learning to Solve Hard Exploration-Exploitation Trade-Offs (NeurIPS 2024 · similarity 0.88)
- Discovering Temporally-Aware Reinforcement Learning Algorithms (ICLR 2024 · similarity 0.87)
- DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- Reason to Play: Behavioral and Brain Alignment Between Frontier LRMs and Human Game Learners (NeurIPS 2026 · similarity 0.85)
- Behavioral Exploration: Learning to Explore via In-Context Adaptation (ICML 2025 · similarity 0.85)
- DynaMITE-RL: A Dynamic Model for Improved Temporal Meta-Reinforcement Learning (NeurIPS 2024 · similarity 0.85)
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.
