Research map / Reinforcement learning / Imitation learning
Maximum-Likelihood Inverse Reinforcement Learning with Finite-Time Guarantees
Siliang Zeng, Chenliang Li, Alfredo Garcia, Mingyi Hong
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 ML-IRL (State-Only) instead of f-IRL | ML-IRL · MuJoCo Hopper | cumulative reward: 3074.55 -> 3089.79 | Improved |
| Use ML-IRL (State-Action) instead of f-IRL | ML-IRL · MuJoCo Ant | cumulative reward: 4571.71 -> 4984.34 | Improved |
| Use ML-IRL (State-Action) instead of f-IRL | ML-IRL · MuJoCo Half-Cheetah | cumulative reward: 4375.88 -> 4086.92 | Hurt |
| Use ML-IRL (State-Only) instead of f-IRL | ML-IRL · MuJoCo Walker | cumulative reward: 4464.20 -> 4380.17 | No clear effect |
| Use ML-IRL (State-Only) instead of f-IRL | ML-IRL · MuJoCo Humanoid | cumulative reward: 5243.90 -> 5390.31 | Improved |
| Use imitation learning BC instead of IRL under a single expert trajectory | BC · MuJoCo Hopper | cumulative reward: 20.49 vs 3089.79 | Hurt |
| Use ML-IRL (State-Only) instead of f-IRL in data transfer | ML-IRL · Custom-Ant to Disabled-Ant data transfer | score: 188.85 -> 221.51 | Improved |
| Use ML-IRL (State-Only) instead of f-IRL in reward transfer | ML-IRL · Custom-Ant to Disabled-Ant reward transfer | score: 156.45 -> 187.69 | Improved |
| Use IQ-Learn instead of f-IRL in transfer | IQ-Learn · Custom-Ant to Disabled-Ant data transfer | score: 188.85 -> -11.78 | Hurt |
| Use state-only reward instead of state-action reward | ML-IRL · MuJoCo Humanoid | cumulative reward: 5390.31 -> 5240.57 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- IQ-Learn — IQ-Learn: Inverse soft-Q Learning for Imitation (NeurIPS 2021) · better on 7 of 7 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Maximum Likelihood Constraint Inference for Inverse Reinforcement Learning (ICLR 2020 · similarity 0.92)
- Recent Advancements in Inverse Reinforcement Learning (AAAI 2024 · similarity 0.92)
- Is Inverse Reinforcement Learning Harder than Standard Reinforcement Learning? A Theoretical Perspective (ICML 2024 · similarity 0.92)
- How does Inverse RL Scale to Large State Spaces? A Provably Efficient Approach (NeurIPS 2024 · similarity 0.91)
- Hybrid Inverse Reinforcement Learning (ICML 2024 · similarity 0.90)
- Offline Inverse RL: New Solution Concepts and Provably Efficient Algorithms (ICML 2024 · similarity 0.90)
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.
