Intrinsic reinforcement learning

Intrinsic Reinforcement Learning

Exploration, play, and curiosity-driven behavior drive the unsupervised synthesis of a broad basis of reusable, primitive skills (mini-policies / macro-actions), which in turn enhances an autonomous agent's ability to solve new problems and attain general competence.

Barto, A.G., Singh, S., and Chentanez, N. (2004) Intrinsically Motivated Learning of Hierarchical Collections of Skills International Conference on Developmental Learning (ICDL), LaJolla, CA, USA pdf

Singh, S., Barto, A.G., and Chentanez, N. (2004) Intrinsically Motivated Reinforcement Learning 18th Annual Conference on Neural Information Processing Systems (NIPS), Vancouver, B.C., Canada, December 2004 pdf

Sutton, R.S., Precup, D., Singh, S. (1999). Between MDPs and semi-MDPs: A Framework for Temporal Abstraction in Reinforcement Learning. Artificial Intelligence 112:181-211. pdf