Generative Adversarial Imitation Learning

Code

• github.com/Kaixhin/imitation-learning
  pytorch★ 563
• github.com/Div99/IQ-Learn
  pytorch★ 377
• github.com/twni2016/f-IRL
  pytorch★ 45
• github.com/ran-weii/cleanil
  pytorch★ 24
• github.com/emunaran/stochastic-human-driving-policies-drl
  pytorch★ 16
• github.com/Techget/gail-tf-sc2
  tf★ 7
• github.com/rohitrango/Reward-bias-in-GAIL
  tf★ 4
• github.com/KshamaDw/collaborative-competitive-games
  tf★ 0
• github.com/nav74neet/gail-tf-gym
  tf★ 0
• github.com/opendilab/DI-engine/blob/main/ding/reward_model/gail_irl_model.py
  pytorch★ 0

Abstract

Consider learning a policy from example expert behavior, without interaction with the expert or access to reinforcement signal. One approach is to recover the expert's cost function with inverse reinforcement learning, then extract a policy from that cost function with reinforcement learning. This approach is indirect and can be slow. We propose a new general framework for directly extracting a policy from data, as if it were obtained by reinforcement learning following inverse reinforcement learning. We show that a certain instantiation of our framework draws an analogy between imitation learning and generative adversarial networks, from which we derive a model-free imitation learning algorithm that obtains significant performance gains over existing model-free methods in imitating complex behaviors in large, high-dimensional environments.

Tasks

Reproductions