Peter Stone's Selected Publications

Classified by TopicClassified by Publication TypeSorted by DateSorted by First Author Last NameClassified by Funding Source


Imitation Learning from Video by Leveraging Proprioception

Faraz Torabi, Garrett Warnell, and Peter Stone. Imitation Learning from Video by Leveraging Proprioception. In Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI), August 2019.

Download

[PDF]1.1MB  [slides.pptx]20.3MB  

Abstract

Classically, imitation learning algorithms have been developed for idealized situations, e.g., the demonstrations are often required to be collected in the exact same environment and usually include the demonstrator's actions. Recently, however, the research community has begun to address some of these shortcomings by offering algorithmic solutions that enable imitation learning from observation (IfO), e.g., learning to perform a task from visual demonstrations that may be in a different environment and do not includeactions. Motivated by the fact that agents often also have access to their own internal states (i.e., proprioception), we propose and study an IfO algorithm that leverages this information in the policy learning process. The proposed architecture learns policies over proprioceptive state representations and compares the resulting trajectories visually to the demonstration data. We experimentally test the proposed technique on several MuJoCo domains and show that it outperforms other imitation from observation algorithms by a large margin.

BibTeX Entry

@InProceedings{IJCAI19b-torabi,
  author = {Faraz Torabi and Garrett Warnell and Peter Stone},
  title = {Imitation Learning from Video by Leveraging Proprioception},
  booktitle = {Proceedings of the 28th International Joint Conference on Artificial Intelligence (IJCAI)},
  location = {Macao, China},
  month = {August},
  year = {2019},
  abstract = {
Classically, imitation learning algorithms have been developed for idealized 
situations, e.g., the demonstrations are often required to be collected in the 
exact same environment and usually include the demonstrator's actions. 
Recently, however, the research community has begun to address some of these 
shortcomings by offering algorithmic solutions that enable imitation learning 
from observation (IfO), e.g., learning to perform a task from visual 
demonstrations that may be in a different environment and do not include
actions. Motivated by the fact that agents often also have access to their own 
internal states (i.e., proprioception), we propose and study an IfO algorithm 
that leverages this information in the policy learning process. The proposed 
architecture learns policies over proprioceptive state representations and 
compares the resulting trajectories visually to the demonstration data. We 
experimentally test the proposed technique on several MuJoCo domains and show 
that it outperforms other imitation from observation algorithms by a large 
margin.
  },
}

Generated by bib2html.pl (written by Patrick Riley ) on Wed Nov 25, 2020 10:07:50