Research map / Reinforcement learning / Policy gradient and actor-critic
Perceiving the Knowledge Boundary: Uncertainty-Guided Exploration and Imagination for World Models
Zhenxian Liu, Peixi Peng, Yangru Huang, Yonghong Tian
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 |
|---|---|---|---|
| add uncertainty-guided exploration and imagination filtering | DreamerV3 · CARLA highway | episode reward: 375 -> 476 | Improved |
| add uncertainty-guided exploration and imagination filtering | DreamerV3 · DMC (6 tasks avg, 1M steps) | return: 456.5 -> 607.9 | Improved |
| add both modules | DreamerV3 · Atari100k Gopher | score: 2977 -> 3122 | Improved |
| model-based method vs model-free | Ours · DMC Finger Spin | return: 756.9 vs DrQ-v2 846.7 | Hurt |
| add exploration only or filtering only (ablation) | DreamerV3 · CARLA highway | reward curve | Improved |
| combine exploration and filtering (ablation) | DreamerV3 · CARLA highway | reward curve | Improved |
| use bootstrap sampling for ensemble (ablation) | ensemble disagreement · CARLA highway | reward curve | No clear effect |
| increase ensemble size from K=3 to 5 (ablation) | ensemble disagreement · CARLA highway | reward curve | Improved |
| increase ensemble size beyond K=5 (ablation) | ensemble disagreement · CARLA highway | reward curve | No clear effect |
| stepwise instead of aggregated uncertainty statistics (ablation) | imagination filtering · CARLA highway | reward curve | Improved |
Closest papers by meaning
- Dynamics-Aligned Latent Imagination in Contextual World Models for Zero-Shot Generalization (NeurIPS 2025 · similarity 0.88)
- Reinforcing VLAs in Task-Agnostic World Models (NeurIPS 2026 · similarity 0.88)
- Open-World Reinforcement Learning over Long Short-Term Imagination (ICLR 2025 · similarity 0.87)
- Beyond Prediction: Steering VLM Agents with Retrospective World Modeling (NeurIPS 2026 · similarity 0.87)
- Beyond One-Size-Fits-All: Diagnosis-Driven Online Reinforcement Learning with Offline Priors (NeurIPS 2026 · similarity 0.87)
- RIG: Synergizing Reasoning and Imagination in End-to-End Generalist Policy (ICLR 2026 · 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.
