Mission
As we navigate through diverse environments and social interactions, we constantly transform sensory information into optimal behaviors. This sensorimotor conversion is inherently complex and non-linear, requiring the brain to flexibly map identical sensory inputs onto a spectrum of potential actions informed by context and immediate goals. Our lab focuses on understanding the computational principles and coding schemes employed by the cortical hierarchy to support this cognitive flexibility and adaptive information processing.
Approach
Our research interests lie at the intersection of systems neuroscience and artificial intelligence, where we integrate complementary computational and experimental approaches. These include deep learning models and machine learning techniques combined with psychophysical and electrophysiological data, such as human intracranial recordings and rodent electrophysiology.
Focus
We also investigate how these computations are disrupted in neuropsychiatric disorders and explore different pathways through which information can be rerouted to bypass disrupted pathways. Additionally, we are excited about the potential of integrating insights on the efficient coding schemes employed by cortical neurons into a new class of biologically-inspired deep learning models. These models, theoretically, would be able to flexibly adapt to changing computational demands while maintaining energy efficiency similar to that of cortical neurons.
The Systems Intelligence Lab is part of the Department of Biomedical Engineering in the Fu Foundation of Engineering and Applied Science at Columbia University. The lab is also affiliated with the Doctoral Program in Neurobiology and Behavior, the Center of Excellence (CoE) in the Neuroscience of Decision-Making, and the Data Science Scholars Program.
Latest
Tomas’ ZIPS presentation
Tomas has been selected to give a talk at the Zuckerman Institute Postdoctoral Seminars (ZIPS) as one of the speakers for Spring/Summer 2024!
Robert and Nuttida at the BCS Young Investigator Lecture Series
Presenting ‘Cortical computations for adaptive learning in humans and artificial networks: Experimental and theoretical approaches’
Nuttida gave a talk at The Manifolds in Nature Conference
Nuttida and Dr. Sejnowski at Okinawa Institute of Science and Technology (OIST)
Joint picnic with the Herbert Wu lab
Herbert Wu lab & Nuttida lab in Central Park for our first joint lab outing!