Research map / Graph learning / GNN architectures
Reward Propagation Using Graph Convolutional Networks
Martin Klissarov, Doina Precup
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 |
|---|---|---|---|
| Use GCN-propagated reward information as a potential for reward shaping | PPO · MiniWorld-FourRooms-v0, MiniWorld-MyWayHome-v0, MiniWorld-MyWayHomeNoisyTv-v0 | score | Improved |
| Use GCN-propagated reward information as a potential for reward shaping | PPO · Atari 2600 games | performance | Improved |
| Use GCN-propagated reward information as a potential for reward shaping | PPO · Delayed MuJoCo environments | performance | Improved |
| Use GCN message passing instead of the exact forward-backward message computation | potential-based reward shaping in FourRooms and FourRoomsTraps · FourRooms and FourRoomsTraps | cumulative steps (regret) | No clear effect |
| Replace the learned GCN potential with an L2 distance-to-goal potential | PPO reward shaping · MiniWorld environments | performance | No clear effect |
| Use RND or ICM exploration shaping instead of the GCN potential | PPO · MiniWorld-MyWayHomeNoisyTv-v0 | performance | Hurt |
| Increase α above 0.5, giving more weight to the regular return (ablation) | Φ_GCN · MiniWorld-FourRooms-v0 | performance | Improved |
| Increase η, weighting propagation loss more strongly (ablation) | Φ_GCN · SMaze-v0 | propagation complexity | Improved |
Closest papers by meaning
- Automatic Reward Shaping from Confounded Offline Data (ICML 2025 · similarity 0.86)
- Bootstrapped Reward Shaping (AAAI 2025 · similarity 0.85)
- How Graph Neural Networks Learn: Lessons from Training Dynamics (ICML 2024 · similarity 0.85)
- STARC: A General Framework For Quantifying Differences Between Reward Functions (ICLR 2024 · similarity 0.85)
- Catching Two Birds with One Stone: Reward Shaping with Dual Random Networks for Balancing Exploration and Exploitation (ICML 2025 · similarity 0.85)
- Highly Efficient Self-Adaptive Reward Shaping for Reinforcement Learning (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.
