Feature learning for multi-task inverse reinforcement learning - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year :

Feature learning for multi-task inverse reinforcement learning

(1) , (1)
1

Abstract

In this paper we study the question of life long learning of behaviors from human demonstrations by an intelligent system. One approach is to model the observed demonstrations by a stationary policy. Inverse rein-forcement learning, on the other hand, searches a reward function that makes the observed policy closed to optimal in the corresponding Markov decision process. This approach provides a model of the task solved by the demonstrator and has been shown to lead to better generalization in un-known contexts. However both approaches focus on learning a single task from the expert demonstration. In this paper we propose a feature learn-ing approach for inverse reinforcement learning in which several different tasks are demonstrated, but in which each task is modeled as a mixture of several, simpler, primitive tasks. We present an algorithm based on an al-ternate gradient descent to learn simultaneously a dictionary of primitive tasks (in the form of reward functions) and their combination into an ap-proximation of the task underlying observed behavior. We illustrate how this approach enables efficient re-use of knowledge from previous demon-strations. Namely knowledge on tasks that were previously observed by the learner is used to improve the learning of a new composite behavior, thus achieving transfer of knowledge between tasks.
Fichier principal
Vignette du fichier
firl.pdf (1.46 Mo) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01098040 , version 1 (14-02-2019)

Licence

Attribution - NonCommercial - NoDerivatives - CC BY 4.0

Identifiers

  • HAL Id : hal-01098040 , version 1

Cite

Olivier Mangin, Pierre-Yves Ouedeyer. Feature learning for multi-task inverse reinforcement learning. 2014. ⟨hal-01098040⟩

Collections

ENSTA INRIA INRIA2
59 View
124 Download

Share

Gmail Facebook Twitter LinkedIn More