Balancing Multiple Sources of Reward in Reinforcement Learning

For many problems which would be natural for reinforcement learning, the reward signal is not a single scalar value but has multiple scalar components. Examples of such problems include agents with multiple goals and agents with multiple users. Creating a single reward value by combining the multiple components can throw away vital information and can lead to incorrect solutions. We describe the multiple reward source problem and discuss the problems with applying traditional reinforcement learning. We then present an new algorithm for finding a solution and results on simulated environments.

• Type: Conference
• Year: 2000
• Where: NIPS
• Authors: Christian R. Shelton
• Keywords: Multiple Scalar Components, NIPS 2000, NIPS 2007, Reinforcement Learning, Single Scalar Value