Research map / Reinforcement learning / Visual and representation RL
Value Propagation Networks
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 |
|---|---|---|---|
| Replace VIN with VProp | VIN · Static grid-world, VIN 16x16 maps | win rate: 63.6% ± 13.2% -> 94.4% ± 5.6% | Helped |
| Replace VIN with MVProp | VIN · Static grid-world, VIN 16x16 maps | win rate: 63.6% ± 13.2% -> 100% | Helped |
| Replace VIN with VProp | VIN · Static grid-world, 32x32 maps | win rate: 15.6% ± 5.3% -> 68.8% ± 27.2% | Helped |
| Replace VIN with MVProp | VIN · Static grid-world, 32x32 maps | win rate: 15.6% ± 5.3% -> 100% | Helped |
| Replace VIN with VProp | VIN · Static grid-world, 64x64 maps | win rate: 4.0% ± 4.1% -> 53.2% ± 31.8% | Helped |
| Replace VIN with MVProp | VIN · Static grid-world, 64x64 maps | win rate: 4.0% ± 4.1% -> 100% | Helped |
| Replace VIN with VProp | VIN · Static grid-world, 32x32 maps | distance to optimal path: 0.8 ± 0.3 -> 0.4 ± 0.3 | Helped |
| Replace VIN with MVProp | VIN · Static grid-world, 32x32 maps | distance to optimal path: 0.8 ± 0.3 -> 0.0 ± .0 | Helped |
| Replace VIN with VProp | VIN · Static grid-world, 64x64 maps | distance to optimal path: 1.5 ± 0.4 -> 0.5 ± 0.4 | Helped |
| Replace VIN with MVProp | VIN · Static grid-world, 64x64 maps | distance to optimal path: 1.5 ± 0.4 -> 0.0 ± .0 | Helped |
| Replace VIN with VProp | VIN · Static grid-world, VIN 16x16 maps | distance to optimal path: 0.2 ± 0.2 -> 0.2 ± 0.2 | No clear effect |
| Replace VIN with MVProp | VIN · Static grid-world, VIN 16x16 maps | distance to optimal path: 0.2 ± 0.2 -> 0.0 ± .0 | Helped |
Closest papers by meaning
- Information-based Value Iteration Networks for Decision Making Under Uncertainty (ICLR 2026 · similarity 0.90)
- Value Iteration Networks (NeurIPS 2016 · similarity 0.89)
- Towards Real-World Navigation With Deep Differentiable Planners (CVPR 2022 · similarity 0.89)
- Scaling Value Iteration Networks to 5000 Layers for Extreme Long-Term Planning (ICML 2025 · similarity 0.88)
- Versatile Navigation Under Partial Observability via Value-guided Diffusion Policy (CVPR 2024 · similarity 0.86)
- Task-Agnostic Pre-training and Task-Guided Fine-tuning for Versatile Diffusion Planner (ICML 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.