Exploiting Multiple Abstractions in Episodic RL via Reward Shaping


  • Roberto Cipollone Università di Roma La Sapienza
  • Giuseppe De Giacomo Università di Roma La Sapienza University of Oxford
  • Marco Favorito Banca d’Italia
  • Luca Iocchi Università di Roma La Sapienza
  • Fabio Patrizi Università di Roma La Sapienza




ML: Reinforcement Learning Theory, ML: Reinforcement Learning Algorithms


One major limitation to the applicability of Reinforcement Learning (RL) to many practical domains is the large number of samples required to learn an optimal policy. To address this problem and improve learning efficiency, we consider a linear hierarchy of abstraction layers of the Markov Decision Process (MDP) underlying the target domain. Each layer is an MDP representing a coarser model of the one immediately below in the hierarchy. In this work, we propose a novel form of Reward Shaping where the solution obtained at the abstract level is used to offer rewards to the more concrete MDP, in such a way that the abstract solution guides the learning in the more complex domain. In contrast with other works in Hierarchical RL, our technique has few requirements in the design of the abstract models and it is also tolerant to modeling errors, thus making the proposed approach practical. We formally analyze the relationship between the abstract models and the exploration heuristic induced in the lower-level domain. Moreover, we prove that the method guarantees optimal convergence and we demonstrate its effectiveness experimentally.




How to Cite

Cipollone, R., De Giacomo, G., Favorito, M., Iocchi, L., & Patrizi, F. (2023). Exploiting Multiple Abstractions in Episodic RL via Reward Shaping. Proceedings of the AAAI Conference on Artificial Intelligence, 37(6), 7227-7234. https://doi.org/10.1609/aaai.v37i6.25881



AAAI Technical Track on Machine Learning I