Aran Nayebi

recurrent neural networks, deep learning, visual neuroscience

I am currently a PhD student in the Neurosciences Graduate Program, co-advised by Surya Ganguli and Dan Yamins. Previously, I completed my M.S. in Computer Science and B.S. in Mathematics & Symbolic Systems at Stanford. My primary interests lie at the intersection of machine learning and neuroscience, where I use tools from deep learning to approach problems in systems neuroscience. Specifically, I am interested in using what we know about the visual system (e.g. the abundance of recurrent connections) to build more biologically realistic models of the ventral visual pathway. I am also interested in using these models to interrogate the role of recurrence in guiding object recognition behavior. My future goals are to use neural network models to produce normative hypotheses about the function and plasticity rules of a neural circuit, as well as closely interact with experimentalists to evaluate these hypotheses and to drive future experiments.