Research map / Label-efficient and robust learning / Self-supervised and contrastive
Latent World Models For Intrinsically Motivated Exploration
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 latent world model prediction error as intrinsic reward | recurrent DQN · Partially Observable Labyrinth 4x4 | final cumulative reward: -1000 -> -23.5 | Helped |
| add latent world model intrinsic reward | recurrent DQN · Partially Observable Labyrinth 5x5 | final cumulative reward: -1000 -> -72.1 | Helped |
| add latent world model intrinsic reward | recurrent DQN · Partially Observable Labyrinth 3x3 | final cumulative reward: -12.4 -> -10.8 | Helped |
| LWM vs best prior method | LWM · Atari Montezuma | final cumulative reward: 387 -> 2276 | Helped |
| LWM vs best prior method | LWM · Atari Gravitar | final cumulative reward: 558 -> 1376 | Helped |
| LWM vs best prior method | LWM · Atari Frostbite | final cumulative reward: 7002 -> 8409 | Helped |
| LWM vs best prior method | LWM · Atari Venture | final cumulative reward: 707 -> 998 | Helped |
| LWM vs best prior method | LWM · Atari Solaris | final cumulative reward: 4467 -> 1268 | Hurt |
| LWM vs best prior method | LWM · Atari Freeway | final cumulative reward: 33.8 -> 30.8 | Hurt |
Closest papers by meaning
- Novelty Search in Representational Space for Sample Efficient Exploration (NeurIPS 2020 · similarity 0.88)
- Zero-shot World Models via Search in Memory (NeurIPS 2025 · similarity 0.87)
- Just Cluster It: An Approach for Exploration in High-Dimensions using Clustering and Pre-Trained Representations (ICML 2024 · similarity 0.87)
- Unlocking the Power of Representations in Long-term Novelty-based Exploration (ICLR 2024 · similarity 0.85)
- Learning with a Mole: Transferable latent spatial representations for navigation without reconstruction (ICLR 2024 · similarity 0.85)
- Open-World Reinforcement Learning over Long Short-Term Imagination (ICLR 2025 · 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.