Amit is a PhD student at Carnegie Mellon University in the Department of Electrical and Computer Engineering. He is advised by Professor Anupam Datta. Prior to this, he received his Bachelor of Technology from the Department of Computer Science and Engineering at Indian Institute of Technology, Kharagpur.
At a high level, Amit's research aims at improving privacy, fairness, and accountability in large-scale automated systems by drawing on techniques in experimentation, statistics, and machine learning. As part of his PhD research, he has developed methods to detect information flow in blackbox systems and applied them to evaluate discrimination, transparency and choice on Google’s ad ecosystem. The information flow experiments framework (AdFisher) is open-sourced and freely available on Github. Details of the experiments and results are available here.
As part of an ongoing project with Microsoft Research, he is evaluating Bing’s big-data advertising pipeline to detect violations of privacy and fairness properties, as well as trace them to specific modules of the pipeline, thereby enabling accountability for violations.