معرفی
Aukosh Jagannath is an Associate Professor in the Department of Statistics and Actuarial Science at the University of Waterloo, with cross-appointments in the Department of Applied Mathematics and the Cheriton School of Computer Science. He is also an affiliate of the Perimeter Institute and Waterloo.AI. His research focuses on probability theory, mathematical foundations of data science, and mathematical physics, particularly in areas like high-dimensional learning, statistical-computational tradeoffs, and spin glasses.
Before joining Waterloo, he held postdoctoral positions at Harvard University and the University of Toronto, and completed his PhD in mathematics at NYU’s Courant Institute under Gerard Ben Arous, alongside an undergraduate degree in math and physics from NYU.
His work has been supported by grants from NSERC, the Canada Research Chairs program, the Canadian Foundation for Innovation, and the Ontario Research Fund. He has contributed to groundbreaking research on topics such as stochastic gradient descent dynamics, spectral alignment in neural networks, and the theoretical underpinnings of deep learning.
Jagannath has advised numerous graduate and postdoctoral researchers, including current PhD students Taj Jones-McCormick and Varnan Sarangian, and former advisees like Aseem Baranwal (now at XTY Labs) and Yiming Xu (now at the University of Kentucky). His awards include an Outstanding Paper Award at NeurIPS 2022 for work on high-dimensional SGD dynamics.
