Research map / Reinforcement learning / Visual and representation RL
TW-CRL: Time-Weighted Contrastive Reward Learning for Efficient Inverse Reinforcement Learning
Yuxuan Li, Yicheng Gao, Ning Yang, Stephen Xia
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 |
|---|---|---|---|
| replace time-weighted rewards with constant +1/-1 rewards (ablation) | TW-CRL · TrapMaze-v2 | return | Hurt |
| replace time-weighted rewards with constant +1/-1 rewards (ablation) | TW-CRL · TrapMaze-v1 | return | Hurt |
| use failed demonstrations in addition to successful ones (ablation) | TW-CRL · TrapMaze-v1 / TrapMaze-v2 | return | Improved |
| add successful demonstrations only (no failed) (ablation) | TW-CRL · TrapMaze-v2 | return | Hurt |
| use TW-CRL reward instead of GAIL/AIRL/MaxEnt reward | TD3 · TrapMaze-v1 | average final return: 253.12 vs 159.06 (GAIL) | Improved |
| use TW-CRL instead of SASR | TD3 · PandaPush | average final return: -96.77 -> -34.89 | Improved |
| use TW-CRL instead of MaxEnt | TD3 · PandaReach | average final return: -5.97 vs -5.98 | No clear effect |
| broaden goal reset from 1 grid to any grid | TW-CRL · TrapMaze-v1 | average return: 253.12 -> 221.33 | Hurt |
| use TW-CRL instead of baselines under broadened goal reset | TD3 · TrapMaze-v1 (any goal grid) | average return: 221.33 vs 122.78 (SASR) | Improved |
| use TW-CRL instead of SASR | TD3 · MountainCar | average final return: 94.23 vs 93.97 | No clear effect |
Closest papers by meaning
- Meta-Inverse Reinforcement Learning with Probabilistic Context Variables (NeurIPS 2019 · similarity 0.89)
- Efficient Exploration of Reward Functions in Inverse Reinforcement Learning via Bayesian Optimization (NeurIPS 2020 · similarity 0.89)
- Recent Advancements in Inverse Reinforcement Learning (AAAI 2024 · similarity 0.88)
- Is Inverse Reinforcement Learning Harder than Standard Reinforcement Learning? A Theoretical Perspective (ICML 2024 · similarity 0.88)
- How does Inverse RL Scale to Large State Spaces? A Provably Efficient Approach (NeurIPS 2024 · similarity 0.88)
- Environment Design for Inverse Reinforcement Learning (ICML 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.
