Research map / Reinforcement learning / Visual and representation RL
Optimistic Exploration in Reinforcement Learning Using Symbolic Model Estimates
Sarath Sreedharan, Michael Katz
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 |
|---|---|---|---|
| bootstrap from lifted model learned on 5x5 grid (ablation) | symbolic exploration with lifting · DoorKey-8x8 | samples: 367749.2 -> 143223.4 | Improved |
| bootstrap from lifted model learned on 5x5 grid (ablation) | symbolic exploration with lifting · DoorKey-8x8 | time (s): 92.78 -> 52.1 | Improved |
| bootstrap from lifted model learned on 5x5 grid (ablation) | symbolic exploration with lifting · DoorKey-6x6 | samples: 21531.8 -> 27666.0 | Hurt |
| bootstrap from lifted model learned on 5x5 grid (ablation) | symbolic exploration with lifting · DoorKey-6x6 | time (s): 35.91 -> 42.14 | Hurt |
| bootstrap from 5x5 grid model (ablation) | symbolic exploration with lifting · DoorKey-16x16 | completion | Improved |
| apply bootstrapping under the original PruneModel strategy (ablation) | symbolic exploration · Minigrid | performance | Hurt |
| replace original PruneModel with strategy reintroducing actions more often (ablation) | symbolic exploration with bootstrapping · DoorKey-8x8 | samples: 260839.8 -> 143223.4 | Improved |
| add lifting generalization (ablation) | symbolic exploration · planning domains | problems solved | Improved |
| use R-max or SMDP exploration instead of symbolic exploration | R-max, SMDP · planning and Minigrid | goal reaching | Hurt |
| initialize Q values from plan found by exploration | tabular Q learning · Minigrid | episode reward | Improved |
| initialize Q values from plan on largest map | tabular Q learning · largest Minigrid | episode reward after 1000 episodes | Improved |
Closest papers by meaning
- Efficient Model-Based Reinforcement Learning Through Optimistic Thompson Sampling (ICLR 2025 · similarity 0.90)
- Beyond Optimism: Exploration With Partially Observable Rewards (NeurIPS 2024 · similarity 0.89)
- SOMBRL: Scalable and Optimistic Model-Based RL (NeurIPS 2025 · similarity 0.89)
- NeoRL: Efficient Exploration for Nonepisodic RL (NeurIPS 2024 · similarity 0.88)
- DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Minimax Optimal Reinforcement Learning with Quasi-Optimism (ICLR 2025 · 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.
