Kush Bhatia

University of California, Berkeley

Kush Bhatia is a PhD student in the EECS Department at the University of California, Berkeley, where he is advised by Peter Bartlett and Anca Dragan.

Kush's research interests broadly lie at the intersection of statistics and machine learning. Currently, he is interested in the design of AI systems whose objectives are well aligned with human values. A key challenge in the design of such systems is to infer underlying human preferences from observed decisions. His research focuses on designing algorithms for this task which are robust to misspecifications, for example, unmodeled biases in human decision making, as well as studying the statistical and computational properties of these procedures.

Kush received his bachelor's degree from Indian Institute of Technology Delhi, majoring in Computer Science and Engineering. After that, he spent a couple of years working at Microsoft Research India as a Research Fellow.