Research map / Reinforcement learning / Imitation learning
Proximal Point Imitation Learning
Luca Viano, Angeliki Kamoutsi, Gergely Neu, Igor Krawczuk, Volkan Cevher
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 P2IL (proximal point) instead of GAIL, AIRL and IQ-Learn in tabular online IL | P2IL · tabular environments (online) | total return | Improved |
| use P2IL with neural network function approximation instead of IQ-Learn | P2IL · CartPole and Acrobot (online) | total return | Hurt |
| use P2IL (dual smoothing) instead of IQ-Learn and AVRIL for offline IL | P2IL offline · offline IL benchmarks | total return | No clear effect |
| run P2IL with image-based observations and neural cost | P2IL · Pong (Atari) | total return | Improved |
| approximate the policy improvement step with SAC for continuous actions | P2IL · MuJoCo Ant, HalfCheetah, Hopper, Walker | total return | Improved |
| transfer the recovered cost to a Gridworld with swapped action effects instead of using the imitating policy | P2IL recovered cost · Gridworld cost transfer | value function / optimality under new dynamics | Improved |
Closest papers by meaning
- Is Behavior Cloning All You Need? Understanding Horizon in Imitation Learning (NeurIPS 2024 · similarity 0.90)
- Imitation Learning in Discounted Linear MDPs without exploration assumptions (ICML 2024 · similarity 0.89)
- Behavior Cloning is Not All You Need: The Optimality of On-Policy Distillation for Noisy Expert Feedback (NeurIPS 2026 · similarity 0.88)
- Learning When to Stop: Selective Imitation Learning Under Arbitrary Dynamics Shift (NeurIPS 2026 · similarity 0.88)
- Beyond-Expert Performance with Limited Demonstrations: Efficient Imitation Learning with Double Exploration (ICLR 2025 · similarity 0.88)
- Zero-Shot Offline Imitation Learning via Optimal Transport (ICML 2025 · 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.
