Research map / Reinforcement learning / Visual and representation RL
Latent State-Predictive Exploration for Deep Reinforcement Learning
Yiming Wang, Kaiyan Zhao, Borong Zhang, Yan Li, Leong Hou U
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 exploration reward function (ERF) (ablation) | LSPE · Stack OOD-A | success rate (%): 31.6% -> 75.7% | Improved |
| add exploration reward function (ERF) (ablation) | LSPE · Stack OOD-C | success rate (%): 17.6% -> 70.9% | Improved |
| add exploration reward function (ERF) (ablation) | LSPE · RealEnv-Hard | success rate (%): 30.3% -> 78.4% | Improved |
| add exploration reward function (ERF) (ablation) | LSPE · RealEnv-Easy | success rate (%): 45.5% -> 91.5% | Improved |
| use temporal-aware encoder instead of MICo-based encoder (ablation) | LSPE · Stack OOD-A | success rate (%): 55.7% -> 75.7% | Improved |
| use temporal-aware encoder instead of MICo-based encoder (ablation) | LSPE · Stack OOD-C | success rate (%): 38.9% -> 70.9% | Improved |
| use temporal-aware encoder instead of MICo-based encoder (ablation) | LSPE · RealEnv-Hard | success rate (%): 49.7% -> 78.4% | Improved |
| LSPE vs best baseline SkillTree on OOD stacking | SkillTree · Stack OOD-A | success rate (%): 44.3% -> 75.7% | Improved |
| LSPE vs best baseline SkillTree on OOD-C | SkillTree · Stack OOD-C | success rate (%): 39.7% -> 70.9% | Improved |
| LSPE vs best baseline EME on real robot hard environment | EME · RealEnv-Hard | success rate (%): 55.4% -> 78.4% | Improved |
| LSPE vs EME on real robot medium environment | EME · RealEnv-Medium | success rate (%): 70.2% -> 85.2% | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SkillTree — SkillTree: Explainable Skill-Based Deep Reinforcement Learning for Long-Horizon Control Tasks (AAAI 2025) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Covariance Volume Maximization for Embodied Latent Exploration in Deep Reinforcement Learning (ICML 2026 · similarity 0.90)
- Random Latent Exploration for Deep Reinforcement Learning (ICML 2024 · similarity 0.89)
- Exploring the Edges of Latent State Clusters for Goal-Conditioned Reinforcement Learning (NeurIPS 2024 · similarity 0.88)
- Imagine Beyond ! Distributionally Robust Autoencoding for State Space Coverage in Online Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- LLM-Explorer: A Plug-in Reinforcement Learning Policy Exploration Enhancement Driven by Large Language Models (NeurIPS 2025 · similarity 0.87)
- Reinforcement Learning Under Latent Dynamics: Toward Statistical and Algorithmic Modularity (NeurIPS 2024 · 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.
