Hosted by the Neural Computation Hub
Abstract: In this talk, I will describe two different projects that investigate the relationship between dynamics and computations during behavior. In the first part of the talk, I will describe a network model for investigating the computational principles underlying the temporal organization of natural behavior in rodents. In the second part of the talk, I will introduce a theory rooted in recurrent networks for accounting for the strong temporal variability but stable encoding observed during delay periods in delayed response tasks in primates.
Bio: Ulises Pereira-Obilinovic is currently a postdoctoral scholar at Xiao-Jing Wang’s lab at New York University (NYU, USA). He will start an independent research position at the Allen Institute for Neural Dynamics (USA) this spring. His research focuses on building theories to understand how behaviour emerges from the interaction of neurons and synapses in the global brain. At NYU, he has worked on problems related to multi-regional computations, reinforcement learning, and the unfolding of natural behavior. He holds two bachelors in Physics and Bioengineering and a master’s in Theoretical Physics from the University of Chile. Under the Fulbright program, he completed a Ph.D. in Statistics at The University of Chicago, supervised by Nicolas Brunel. In his dissertation, he studied learning and memory in recurrent network models using statistical physics.