Research map / Reinforcement learning / Visual and representation RL
Learning to Utilize Shaping Rewards: A New Approach of Reward Shaping
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 shaping rewards (any shaping method) to PPO | PPO · cartpole discrete-action | ASPE: 170 -> 200 | Helped |
| learn adaptive shaping weights (BiPaRS) instead of naive shaping/DPBA with harmful shaping reward | PPO · cartpole harmful shaping | ASPE | Helped |
| use BiPaRS-MGL/IMGL instead of NS, DPBA, RCPO | PPO · MuJoCo Hopper-v2 with torque shaping | ARPE | Helped |
| use explicit mapping (EM) instead of meta-gradient (MGL/IMGL) gradient approximation (ablation) | BiPaRS · MuJoCo Hopper-v2 | ARPE | Hurt |
| reload shaping weight function learnt in first test (ablation) | BiPaRS · cartpole harmful shaping | ASPE | Helped |
| use state-dependent shaping weight function instead of a single shaping weight (ablation) | BiPaRS · cartpole uniformly beneficial/harmful shaping | ASPE | Hurt |
| use state-dependent shaping weight function instead of a single shaping weight (ablation) | BiPaRS · cartpole half-helpful half-harmful shaping | ASPE | Helped |
Closest papers by meaning
- Automatic Reward Shaping from Confounded Offline Data (ICML 2025 · similarity 0.91)
- Highly Efficient Self-Adaptive Reward Shaping for Reinforcement Learning (ICLR 2025 · similarity 0.91)
- Reward Shaping for Reinforcement Learning with An Assistant Reward Agent (ICML 2024 · similarity 0.91)
- ORSO: Accelerating Reward Design via Online Reward Selection and Policy Optimization (ICLR 2025 · similarity 0.91)
- BAMDP Shaping: a Unified Framework for Intrinsic Motivation and Reward Shaping (ICLR 2025 · similarity 0.90)
- Centralized Reward Agent for Knowledge Sharing and Transfer in Multi-Task Reinforcement Learning (NeurIPS 2025 · similarity 0.89)
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.