Babak Hassibi

Babak Hassibi was born in 1970 in Tehran, Iran, and grew up during a period of profound social transformation in his country. He demonstrated exceptional mathematical talent from an early age and pursued his undergraduate studies in electrical engineering at the Sharif University of Technology in Tehran, one of Iran's most prestigious technical institutions. He subsequently emigrated to the United States, where he earned his Master's degree and PhD from Stanford University, working under the supervision of Thomas Kailath, a giant in the field of signal processing and information theory.

After completing his doctorate, Hassibi joined Bell Laboratories, then the world's premier industrial research institution, where he worked alongside some of the foremost minds in communications and information theory. His research during this period produced foundational results on MIMO wireless communications, including the derivation of capacity bounds for multiple-input multiple-output antenna systems that became cornerstones of 3G, 4G, and subsequent wireless standards. His analysis provided the mathematical scaffolding that enabled modern mobile broadband.

Hassibi joined the faculty of the California Institute of Technology (Caltech) in 2001, where he became a full professor in the Department of Electrical Engineering. At Caltech, his research broadened to encompass compressed sensing, random matrix theory, optimization, and more recently the theoretical foundations of deep learning and neural networks. He has published landmark papers on the implicit bias of gradient descent in overparameterized models, contributing fundamental understanding to why modern neural networks generalize well despite their enormous size.

Hassibi has received numerous awards including the IEEE Kiyo Tomiyasu Award and the Okawa Foundation Research Award. He is a Fellow of the IEEE. He has supervised dozens of PhD students who have gone on to leading academic and industrial positions worldwide. He remains one of the most mathematically rigorous theorists at the interface of information theory, signal processing, and machine learning.