Imitation Learning from Observations by Minimizing Inverse Dynamics Disagreement

by   Chao Yang, et al.

This paper studies Learning from Observations (LfO) for imitation learning with access to state-only demonstrations. In contrast to Learning from Demonstration (LfD) that involves both action and state supervision, LfO is more practical in leveraging previously inapplicable resources (e.g. videos), yet more challenging due to the incomplete expert guidance. In this paper, we investigate LfO and its difference with LfD in both theoretical and practical perspectives. We first prove that the gap between LfD and LfO actually lies in the disagreement of inverse dynamics models between the imitator and the expert, if following the modeling approach of GAIL. More importantly, the upper bound of this gap is revealed by a negative causal entropy which can be minimized in a model-free way. We term our method as Inverse-Dynamics-Disagreement-Minimization (IDDM) which enhances the conventional LfO method through further bridging the gap to LfD. Considerable empirical results on challenging benchmarks indicate that our method attains consistent improvements over other LfO counterparts.



page 8

page 21

page 23


Off-Policy Imitation Learning from Observations

Learning from Observations (LfO) is a practical reinforcement learning s...

State-Only Imitation Learning for Dexterous Manipulation

Dexterous manipulation has been a long-standing challenge in robotics. R...

To Follow or not to Follow: Selective Imitation Learning from Observations

Learning from demonstrations is a useful way to transfer a skill from on...

RIDM: Reinforced Inverse Dynamics Modeling for Learning from a Single Observed Demonstration

Imitation learning has long been an approach to alleviate the tractabili...

On the Guaranteed Almost Equivalence between Imitation Learning from Observation and Demonstration

Imitation learning from observation (LfO) is more preferable than imitat...

Optimism is All You Need: Model-Based Imitation Learning From Observation Alone

This paper studies Imitation Learning from Observations alone (ILFO) whe...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.