Arman Afrasiyabi

I am an Assistant Professor in the Department of Neurosurgery at Yale University. My research develops interpretable multimodal AI, foundation-model, and time-series methods for complex neural and biomedical data. Neural activity is inherently spatiotemporal, unfolding across anatomy and across time at once. I design models that learn both of those axes together and align many kinds of signal, so that brain activity can be decoded reliably across individuals, cognitive tasks, and the different environments in which recordings are made.

Before joining the Yale faculty, I held research appointments in the Departments of Computer Science, Genetics, and Neurosurgery at Yale, where I worked on brain decoding and designed biologically grounded multimodal transformers for genomic data. I completed my doctoral work at Université Laval and Mila–Quebec AI Institute, at the crossing point of artificial intelligence, representation learning, and meta-learning.