Achievement

The second of two bootstrapping graduate courses in Computer Science, Computational Modeling was offered in spring. The course drew on a range of research projects carried out by perceptual science faculty to develop a closer link between class topics in probabilistic reasoning and research projects in perceptual science. In teaching randomized methods for probabilistic reasoning, the material was linked to research on ambiguities of memory retrieval (Hemmer) and ambiguities of perceptual grouping (Feldman, Singh). A new class project explored similar ideas in the domain of image segmentation (as studied by Elgammal and Meer), which provided links not only to probabilistic reasoning, but also to models of scene structure and image formation. The resulting material taught relevant concepts, mathematics and algorithms linked with the problems of perception and opportunities for further research by perceptual science students.