Clustering with Reinforcement Learning
- 2.1k Downloads
We show how a previously derived method of using reinforcement learning for supervised clustering of a data set can lead to a sub-optimal solution if the cluster prototypes are initialised to poor positions. We then develop three novel reward functions which show great promise in overcoming poor initialization. We illustrate the results on several data sets. We then use the clustering methods with an underlying latent space which enables us to create topology preserving mappings. We illustrate this method on both real and artificial data sets.
KeywordsLatent Point Reinforcement Learning Reward Function Bernoulli Model Supervise Cluster
Unable to display preview. Download preview PDF.
- 1.Barbakh, W.: The family of inverse exponential k-means algorithms. Computing and Information Systems 11(1), 1–10 (2007)Google Scholar
- 3.Barbakh, W., Fyfe, C.: Performance functions and clustering algorithms. Computing and Information Systems 10(2), 2–8 (2006)Google Scholar
- 4.Barbakh, W., Fyfe, C.: Tailoring local and global interactions in clustering algorithms. Technical Report 40, School of Computing, University of Paisley (March 2007), ISSN 1461-6122Google Scholar
- 5.Bishop, C.M., Svensen, M., Williams, C.K.I.: Gtm: The generative topographic mapping. Neural Computation (1997)Google Scholar
- 7.Kaelbling, L.P., Littman, M.L., Moore, A.W.: Reinforcement learning: A survey. Journal of Artificial Intelligence Research 4, 237–285 (1996)Google Scholar
- 8.Kohonen, T.: Self-Organising Maps. Springer, Heidelberg (1995)Google Scholar
- 9.Likas, A.: A reinforcement learning approach to on-line clustering. Neural Computation (2000)Google Scholar
- 10.Sutton, R.S., Barto, A.G.: Reinforcement Learning: an Introduction. MIT Press, Cambridge (1998)Google Scholar