Subhankar Mishra
Grant Category: Fulbright-Nehru Academic & Professional Excellence Award (Research & Teaching)
Project Title: Secure, Private, and Fair Machine Learning for Insider Threat Detection
Field of Study: Computer Sciences
Home Institution: National Institute of Science Education and Research, Bhubaneswar, Odisha
Host Institution: University of Florida, Gainesville, FL  
Grant Start Month: January, 2022
Duration of Grant: Seven months

Subhankar Mishra
Brief Bio:

Dr. Subhankar Mishra is a Reader-F in the School of Computer Sciences, NISER, Bhubaneswar, Odisha. He has been working on machine learning with a special interest in security and privacy in machine learning. In the past few years, he has also been developing novel methods to make the privacy in machine learning practical. His other areas of interest include graph neural networks, 3D modeling, and low-resource natural language processing.

Dr. Mishra completed his PhD at the University of Florida in 2016 and his BTech at NIT Rourkela in 2010. He joined Oak Ridge National Lab as a research associate to study cyberattacks on Smart Grids. His projects in collaboration with India’s Department of Science and Technology involve automating the development of building information models through machine learning. Another project with La Fondation Dassault Systèmes aims to conserve historical monuments with Heritage BIM. Additionally, using machine learning, he aims to look for indicators and factors affecting nutritional deficiency in India and other nations through the BRICS project, which materialized after getting selected for BRICS YSF 2018.

Dr. Mishra’s Fulbright-Nehru project aims to detect insider threats while preserving the privacy of the users. His goal is to understand and implement machine learning algorithms that are fair to all users and leverage community and group dynamics in understanding and detecting anomalies. Dr. Mishra is team teaching a course on secure and private machine learning dealing in part with the protection of sensitive data while implementing machine learning algorithms.