ICML 2011 workshop on unsupervised and transfer learning

Transfer Learning in Sequential Decision Problems:
A Hierarchical Bayesian Approach
Aaron Wilson, Alan Fern, and Prasad Tadepalli
Oregon State University

Transfer learning is one way to close the gap between the apparent speed of learning of humans in novel domains and the relatively slow pace of learning of machines. Transfer is doubly beneficial in reinforcement learning where the agent not only needs to generalize from sparse experience, but also needs to discover good opportunities to learn in the first place. In this paper, we show that the hierarchical Bayesian framework can be readily adapted to sequential decision problems and provides a natural formalization of transfer learning. We present positive transfer results in learning action models and policies in this framework through empirical evaluations in real-time strategy games.