Compatible Reward Inverse Reinforcement Learning

Alberto Metelli 1 Matteo Pirotta 2 Marcello Restelli 1
2 SEQUEL - Sequential Learning
Inria Lille - Nord Europe, CRIStAL - Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189
Abstract : Inverse Reinforcement Learning (IRL) is an effective approach to recover a reward function that explains the behavior of an expert by observing a set of demonstrations. This paper is about a novel model-free IRL approach that, differently from most of the existing IRL algorithms, does not require to specify a function space where to search for the expert's reward function. Leveraging on the fact that the policy gradient needs to be zero for any optimal policy, the algorithm generates a set of basis functions that span the subspace of reward functions that make the policy gradient vanish. Within this subspace, using a second-order criterion, we search for the reward function that penalizes the most a deviation from the expert's policy. After introducing our approach for finite domains, we extend it to continuous ones. The proposed approach is empirically compared to other IRL methods both in the (finite) Taxi domain and in the (continuous) Linear Quadratic Gaussian (LQG) and Car on the Hill environments.
Document type :
Conference papers
Complete list of metadatas

Cited literature [39 references]  Display  Hide  Download

https://hal.inria.fr/hal-01653328
Contributor : Alessandro Lazaric <>
Submitted on : Friday, December 1, 2017 - 12:17:50 PM
Last modification on : Tuesday, August 13, 2019 - 11:10:04 AM

File

6800-compatible-reward-inverse...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01653328, version 1

Citation

Alberto Metelli, Matteo Pirotta, Marcello Restelli. Compatible Reward Inverse Reinforcement Learning. The Thirty-first Annual Conference on Neural Information Processing Systems - NIPS 2017, Dec 2017, Long Beach, United States. ⟨hal-01653328⟩

Share

Metrics

Record views

276

Files downloads

144