Maximum Entropy Semi-Supervised Inverse Reinforcement Learning

Julien Audiffren 1 Michal Valko 2 Alessandro Lazaric 2 Mohammad Ghavamzadeh 2
2 SEQUEL - Sequential Learning
Inria Lille - Nord Europe, CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189
Abstract : A popular approach to apprenticeship learning (AL) is to formulate it as an inverse reinforcement learning (IRL) problem. The MaxEnt-IRL algorithm successfully integrates the maximum entropy principle into IRL and unlike its predecessors, it resolves the ambiguity arising from the fact that a possibly large number of policies could match the expert's behavior. In this paper, we study an AL setting in which in addition to the expert's trajectories, a number of unsupervised trajectories is available. We introduce MESSI, a novel algorithm that combines MaxEnt-IRL with principles coming from semi-supervised learning. In particular, MESSI integrates the unsupervised data into the MaxEnt-IRL framework using a pairwise penalty on trajectories. Empirical results in a highway driving and grid-world problems indicate that MESSI is able to take advantage of the unsupervised trajectories and improve the performance of MaxEnt-IRL.
Complete list of metadatas

Cited literature [8 references]  Display  Hide  Download

https://hal.inria.fr/hal-01146187
Contributor : Alessandro Lazaric <>
Submitted on : Monday, July 20, 2015 - 10:10:21 AM
Last modification on : Friday, March 22, 2019 - 1:35:14 AM
Long-term archiving on: Wednesday, October 21, 2015 - 5:00:43 PM

File

messi-TR.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01146187, version 1

Citation

Julien Audiffren, Michal Valko, Alessandro Lazaric, Mohammad Ghavamzadeh. Maximum Entropy Semi-Supervised Inverse Reinforcement Learning. International Joint Conference on Artificial Intelligence, Jul 2015, Bueons Aires, Argentina. ⟨hal-01146187⟩

Share

Metrics

Record views

724

Files downloads

424