Belief Reward Shaping in Reinforcement Learning

Abstract

A key challenge in many reinforcement learning problems is delayed rewards, which can significantly slow down learning. Although reward shaping has previously been introduced to accelerate learning by bootstrapping an agent with additional information, this can lead to problems with convergence. We present a novel Bayesian reward shaping framework that augments the reward distribution with prior beliefs that decay with experience. Formally, we prove that under suitable conditions a Markov decision process augmented with our framework is consistent with the optimal policy of the original MDP when using the Q-learning algorithm. However, in general our method integrates seamlessly with any reinforcement learning algorithm that learns a value or action-value function through experience. Experiments are run on a gridworld and a more complex backgammon domain that show that we can learn tasks significantly faster when we specify intuitive priors on the reward distribution.

Ofir Marom

My PhD research focused on reducing learning complexity in Reinforcement Learning and Planning agents by leveraging prior knowledge and transferring knowledge from previous tasks.

Benjamin Rosman

Lab Director

I am a Professor in the School of Computer Science and Applied Mathematics at the University of the Witwatersrand in Johannesburg. I work in robotics, artificial intelligence, decision theory and machine learning.