Minimal Mathematical Models For Human Categorization

Faculty Advisors: Rosa Arriaga and Santosh Vempala

Areas of Research: Algorithms, Cognition.

Compensation: Credit or Pay.

Job Description:
How do humans learn to categorize complex objects from very few examples? The goal of this project is to test the hypothesis that the brain performs random projection as a preprocessing step to obtain a low-dimensional representation. Students will design stimuli and conduct experiments with humans and neural networks. The results will be used to refine and extend theories of cognitive categorization.

News: David Mann and David Rutter won first prize for an oral presentation at the institute research symposium and second prize for their poster at the College of Computing undergrad research symposium!!

Contact email: or