Research map / Reinforcement learning / Visual and representation RL
Learning Task-Distribution Reward Shaping with Meta-Learning
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 meta-learned potential-based reward shaping instead of naive task averaging | MD3QN · CartPoles | return | Helped |
| Use meta-learned potential-based reward shaping instead of naive task averaging | MD3QN · Grid mazes | return | Helped |
| Use meta-learned potential-based reward shaping instead of naive task averaging | MD3QN · CoinRun | solve ratio | Helped |
| Add learned reward shaping to vanilla MAML adaptation (ablation) | MD3QN on CartPoles, grid mazes, and CoinRun · CartPoles, grid mazes, and CoinRun | return | Helped |
| Add a dueling value-and-advantage architecture (ablation) | MD3QN on CartPoles, grid mazes, and CoinRun · CartPoles, grid mazes, and CoinRun | return | Helped |
| Replace the dueling architecture with MDDQN (ablation) | Meta-learned reward shaping · CartPoles, grid mazes, and CoinRun | return | Hurt |
| Use learned shaping instead of no shaping | RL agents on grid mazes · Grid mazes | return | Helped |
| Replace learned shaping with random noise shaping (ablation) | RL agents on grid mazes · Grid mazes | return | Hurt |
| Replace learned shaping with a randomly initialized network (ablation) | RL agents on grid mazes · Grid mazes | return | Hurt |
| Use learned shaping instead of Manhattan- or Euclidean-distance shaping | RL agents · Grid mazes | return | Helped |
| Use the learned potential function when transferring from discrete-action training to continuous-action testing | DDPG agents · CartPoles with continuous actions | return | Helped |
| Add learned reward shaping to PEARL (ablation) | PEARL · CoinRun | solve ratio | Helped |
Closest papers by meaning
- Highly Efficient Self-Adaptive Reward Shaping for Reinforcement Learning (ICLR 2025 · similarity 0.90)
- Action-Dependent Optimality-Preserving Reward Shaping (ICML 2025 · similarity 0.89)
- Reward Shaping for Reinforcement Learning with An Assistant Reward Agent (ICML 2024 · similarity 0.89)
- DISCOVER: Automated Curricula for Sparse-Reward Reinforcement Learning (NeurIPS 2025 · similarity 0.88)
- Automatic Reward Shaping from Confounded Offline Data (ICML 2025 · similarity 0.88)
- Potential-Based Reward Shaping for Intrinsic Motivation (Student Abstract) (AAAI 2024 · similarity 0.88)
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.