Reinforcement Learning with Stochastic Reward Machines


  • Jan Corazza University of Zagreb Max Planck Institute for Software Systems
  • Ivan Gavran Max Planck Institute for Software Systems
  • Daniel Neider Max Planck Institute for Software Systems



Machine Learning (ML)


Reward machines are an established tool for dealing with reinforcement learning problems in which rewards are sparse and depend on complex sequences of actions. However, existing algorithms for learning reward machines assume an overly idealized setting where rewards have to be free of noise. To overcome this practical limitation, we introduce a novel type of reward machines, called stochastic reward machines, and an algorithm for learning them. Our algorithm, based on constraint solving, learns minimal stochastic reward machines from the explorations of a reinforcement learning agent. This algorithm can easily be paired with existing reinforcement learning algorithms for reward machines and guarantees to converge to an optimal policy in the limit. We demonstrate the effectiveness of our algorithm in two case studies and show that it outperforms both existing methods and a naive approach for handling noisy reward functions.




How to Cite

Corazza, J., Gavran, I., & Neider, D. (2022). Reinforcement Learning with Stochastic Reward Machines. Proceedings of the AAAI Conference on Artificial Intelligence, 36(6), 6429-6436.



AAAI Technical Track on Machine Learning I