Research map / Reinforcement learning / Imitation learning
Error Bounds of Imitating Policies and Environments
Tian Xu, Ziniu Li, Yang Yu
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 |
|---|---|---|---|
| Match state-action occupancy distributions with GAIL instead of cloning actions with BC | Policy imitation · Infinite-horizon MDPs | policy value gap bound: BC: 1/(1-γ)^2; GAIL: 1/(1-γ) | Improved |
| Learn the transition model with a GAIL-style distribution-matching objective instead of one-step BC | Environment imitation · Policy evaluation in learned environments | model-bias term in policy evaluation error bound: BC: 1/(1-γ)^2; GAIL: 1/(1-γ) | Improved |
| Use GAIL rather than BC to learn the transition model | Environment imitation · Three MuJoCo environments | policy evaluation error | Improved |
| Use gradient penalty with a moderate coefficient rather than no penalty or a very large coefficient (ablation) | GAIL · MuJoCo policy imitation | performance | Improved |
| Reduce the number of discriminator network layers (ablation) | GAIL · MuJoCo policy imitation | performance | No clear effect |
| Provide BC with many expert demonstrations rather than scant demonstrations | BC · MuJoCo policy imitation | performance relative to expert | Improved |
| Use BC rather than GAIL as the transition-model learner | Environment imitation · HalfCheetah-v2 | policy evaluation error during learning | Hurt |
Closest papers by meaning
- Error Bounds of Imitating Policies and Environments for Reinforcement Learning (TPAMI 2021 · similarity 0.99)
- Generative Adversarial Imitation Learning (NeurIPS 2016 · similarity 0.90)
- Behavioral Cloning from Noisy Demonstrations (ICLR 2021 · similarity 0.90)
- Sample Efficient Imitation Learning for Continuous Control (ICLR 2019 · similarity 0.90)
- EvIL: Evolution Strategies for Generalisable Imitation Learning (ICML 2024 · similarity 0.88)
- Diffusion Imitation from Observation (NeurIPS 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.
