Brian DePasquale

Brian DePasquale, PhD, is an Assistant Professor in the Department of Biomedical Engineering, and a Hariri Institute Faculty Affiliate (2023) at Boston University. He conducts research in theoretical neuroscience and machine learning. His research uses mathematical models to characterize and explain how populations of neurons perform computations to produce behavior. Presently, he works with Carlos Brody and Jonathan Pillow on latent variable models of evidence accumulation. During his Ph.D. he worked with Larry Abbott developing methods for training recurrent neural networks, and with Mark Churchland connecting these models conceptually to statistical models of low-dimensional dynamics applied to data. Before that he studied the basal ganglia in the laboratory of Ann Graybiel.

In addition to doing science, he sometimes write about other people’s science, and also tries to contribute to open-source computing libraries.

Google Scholar Profile