A Predictive Model for Imitation Learning in Partially Observable Environments

A Boularias

doi:10.1109/ICMLA.2008.142

Back

A Predictive Model for Imitation Learning in Partially Observable Environments

Conference proceeding

Open access

A Predictive Model for Imitation Learning in Partially Observable Environments

A Boularias

2008 Seventh International Conference on Machine Learning and Applications, pp.83-90

12/2008

DOI: https://doi.org/10.1109/ICMLA.2008.142

Abstract

Automatic control

Human robot interaction

Imitation Learning

Intelligent control

Intelligent robots

Machine learning

POMDPs

Power system modeling

Predictive models

PSRs

Stochastic systems

Switches

Wheelchairs

Learning by imitation has shown to be a powerful paradigm for automated learning in autonomous robots. This paper presents a general framework of learning by imitation for stochastic and partially observable systems. The model is a Predictive Policy Representation (PPR) whose goal is to represent the teacher's policies without any reference to states. The model is fully described in terms of actions and observations only. We show how this model can efficiently learn the personal behavior and preferences of an assistive robot user.

Files and links (1)

url

https://doi.org/10.1109/ICMLA.2008.142View

Version of Record (VoR) Open

Metrics

6 Record Views

Details

Title: A Predictive Model for Imitation Learning in Partially Observable Environments
Creators: A Boularias - Dept. of Comput. Sci., Laval Univ., Montreal, QC, Canada
Publication Details: 2008 Seventh International Conference on Machine Learning and Applications, pp.83-90
Date published: 12/2008
Publisher: IEEE
Academic Unit: Computer Science (SAS)
Language: English
Resource Type: Conference proceeding
Identifiers: 991031665837404646