Summary
Sushrut Thorat is a postdoctoral researcher and computational cognitive neuroscientist with a decade of experience probing how brains and artificial systems learn, particularly under non-i.i.d. streaming conditions. Based at Universität Osnabrück, he develops neuroconnectionist models of visual representations and sequential learning informed by normative modeling and neuroimaging-derived priors. His PhD work at the Donders Institute combined behavioral experiments, fMRI, and machine learning to dissect attentional modulation and visual statistical learning in humans and networks. Trained in engineering physics (IIT Bombay) and cognitive neuroscience (Università di Trento, cum laude), he blends rigorous quantitative skills with cognitive theory to design non-traditional neural architectures tailored to task constraints. He is equally interested in the early developmental trajectory of human brains, bringing developmental perspectives to the engineering of more robust, lifelong-learning AI.
10 years of coding experience
4 years of employment as a software developer
Doctor of Philosophy, Cognitive Neuroscience, Doctor of Philosophy, Cognitive Neuroscience at Radboud University
Master of Science, Cognitive Science, Master of Science, Cognitive Science at Università di Trento
Indian Institute of Technology Bombay