Selected Recent Publications:

PhD Thesis

• Hierarchical Reinforcement Learning: A Hybrid Approach, Ryan, M.R.K. PhD Thesis, University of NSW, School of Computer Science and Engineering, 2004. Click to download abstract, full paper in gzipped postscript (662K) or gzipped PDF (1.1M).

Book chapters

• Hierarchical Decision Making, Ryan, M.R.K. In Handbook of Learning and Approximate Dynamic Programming, J. Si, A Barto, W. Powell, D Wunsch (Eds), IEEE Press, 2004. Preprint

Journal papers

Conference papers

• Using Abstract Models of Behaviours to Automatically Generate Reinforcement Learning Hierarchies., Ryan, M.R.K, In Proceedings of The 19th International Conference on Machine Learning, Sydney, Australia, July 2002. Click to download abstract or full paper (gzipped postscript)
• Learning to Fly: An Application of Hierarchical Reinforcement Learning., Ryan, M.R.K, Reid M.D., In Proceedings of The 17th International Conference on Machine Learning, San Francisco, California, June 2000. Click to download abstract or full paper (compressed postscript)
• Using ILP to Improve Planning in Hierarchical Reinforcement Learning., Reid M.D., Ryan, M.R.K, In Proceedings of the Tenth International Conference on Inductive Logic Programming, London, UK, July 2000. Click to download abstract or full paper (compressed postscript)
• RL-TOPs: An Architecture for Modularity and Re-Use in Reinforcement Learning., Ryan, M.R.K, Pendrith M.D., In Proceedings of The 15th International Conference on Machine Learning, Madison, Wisconsin, July 1998. Click to download abstract or full paper (compressed postscript)
• Estimator variance in reinforcement learning: Theoretical problems and practical solutions, Pendrith M.D., Ryan M.R.K., AAAI-97 Workshop on On-line Search, Providence, RI, 28 July 1997. Click to download abstract or full paper (compressed postscript).
• Actual return reinforcement learning versus Temporal Differences: Some theoretical and experimental results, Pendrith M.D., Ryan M.R.K., The 13th International Conference on Machine Learning, Bari, Italy, 3-6 July 1996. Click to download abstract or full paper (compressed postscript).
• C-Trace: A new algorithm for reinforcement learning of robotic control., Pendrith M.D., Ryan M.R.K., ROBOLEARN-96, Key West, Florida, 19-20 May, 1996. Click to download abstract or full paper (compressed postscript)