Paul W. Munro Associate Professor, SIS-Information Science & Telecommunications University of Pittsburgh Email: pmunro@mail.sis.pitt.edu Individual Website: http://www.sis.pitt.edu/~pmunro

Ph. D., Brown University Research Interests My primary research interest is to abstract mathematical and computational principles underlying learning at the synaptic, neuronal, and systems levels. Thus, I have conducted studies that examine neurophysiological phenomena such as the development of feature sensitive neurons in sensory cortical areas, long term potentiation (LTP), and most recently spike-timing dependent plasticity (STDP). Ultimately, I feel that neural network models provide a language for understanding the relation between the mind (cognitive phenomena) and the brain (neurobiological). Thus I also have a strong interest in developing neurally inspired models at a cognitive level. Projects at this level include: The acquisition of spatial information to form a cognitive map. Analysis of the impact of early learning on adult representations learning in the absense of a teacher (self-supervised learning)transfer of learning from a previously learned task to a novel task. A final component to my research agenda is the application of neural learning processes to real-world problems. Neural network learning procedures have resulted in novel approaches to tasks in business, engineering, and medicine. Togther with my students and colleagues, I have developed approaches to the following set of tasks: Image Compression (data storage and communication) Medical Diagnosis (medicine) Treatment Optimization (medicine) Inference of personal data from television viewing habits (marketing) Recent Publications P. W. Munro and G. Cottrell (2001) Connectionist Network Interactions between Frequency Effects and Age of Acquisition Effects in a Connectionist Network. Twenty Third Ann. Conf. Cognitive Science Society Proceedings. pp. 703-708. Lawerence Erlbaum: Mahwah NJ. P. Munro and S. Sanguansintukual (2002) A Neural Network Approach to Treatment Optimization, AMIA 2002 Symposium Proceedings, American Medical Informatics Association. G. Hernandez, P. Munro, and J. Rubin (2003) The effect of spike redistribution in a reciprocally connected pair of neurons wth spike-timing dependent plasticity. Neurocomputing. 52: 347-353. P. W. Munro and G. Hernandez (2000) LTD facilitates learning in a noisy environment. In: S. A. Solla, T. K. Leen, K-R. Muller, eds. Advances in Neural Information Processing Systems 12. MIT Press: Cambridge, MA. T. Ghiselli-Crippa and P. W. Munro (2000) Effects of spatial and temporal contiguity on the acquisition of spatial information. In: S. A. Solla, T. K. Leen, K-R. Muller, eds. . Advances in Neural Information Processing Systems 12. MIT Press: Cambridge, MA. P. W. Munro and B. Parmanto (1997) Competition among networks improves committee performance. In: M. C. Mozer, M. I. Jordan, T. Petsche, eds. Advances in Neural Information Processing Systems 9. MIT Press: Cambridge, MA. B. Parmanto, P. W. Munro, and H. R. Doyle (1996) Reducing variance of committee prediction with resampling techniques. Connection Science. 8: 405-425.