Research map / Probabilistic and scientific modeling / Causal discovery
Schema Networks: Zero-shot Transfer with a Generative Causal Model of Intuitive Physics
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 |
|---|---|---|---|
| Use Schema Networks instead of A3C Image + Entities for zero-shot transfer | Breakout agents · Offset Paddle | average score per episode: 11.10 ± 17.44 -> 41.42 ± 6.29 | Helped |
| Use Schema Networks instead of A3C Image Only for zero-shot transfer | Breakout agents · Middle Wall | average score per episode: 9.55 ± 17.44 -> 35.22 ± 12.23 | Helped |
| Use Schema Networks instead of A3C Image + Entities for zero-shot transfer | Breakout agents · Random Target | average score per episode: 6.88 ± 6.19 -> 21.38 ± 5.02 | Helped |
| Use Schema Networks instead of A3C Image + Entities for zero-shot transfer | Breakout agents · Juggling | average score per episode: -17.52 ± 17.39 -> -0.11 ± 0.34 | Helped |
| Use Schema Networks instead of A3C Image + Entities | Breakout agents · Half Negative Bricks | average score per episode: 1.55 ± 6.41 -> 5.77 ± 4.30 | Helped |
| Add entity channels to the A3C image input | A3C · Offset Paddle | average score per episode: 0.60 ± 20.05 -> 11.10 ± 17.44 | Mixed |
| Add entity channels to the A3C image input | A3C · Middle Wall | average score per episode: 9.55 ± 17.44 -> 8.00 ± 14.61 | No clear effect |
Closest papers by meaning
- Differentiable Learning of Lifted Action Schemas for Classical Planning (NeurIPS 2026 · similarity 0.86)
- Kinetix: Investigating the Training of General Agents through Open-Ended Physics-Based Control Tasks (ICLR 2025 · similarity 0.85)
- Rapidly Adapting Policies to the Real-World via Simulation-Guided Fine-Tuning (ICLR 2025 · similarity 0.85)
- A Fixed-Point Approach for Causal Generative Modeling (ICML 2024 · similarity 0.85)
- Modeling Unseen Environments with Language-guided Composable Causal Components in Reinforcement Learning (ICLR 2025 · similarity 0.85)
- ReGen: Generative Robot Simulation via Inverse Design (ICLR 2025 · similarity 0.84)
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.