Talk
Talk by Asif Khan
Title: Dynamic Planar Graph Isomorphism is in DynFO Teams link
Abstract:
Machine Learning systems are now ubiquitous in many critical domains such as healthcare and finance. Therefore, practitioners are now interested in enforcing several Responsible AI constraints in the training of such models. In this talk, we will cover one such class of constraints, fairness. We will discuss common metrics, implementation strategies, and the limitations of various fairness metrics and algorithms. Towards the end, we shall also discuss some extensions to this problem, such as approaches to quantify data fairness, mitigate selection bias, and address incentive-driven behavior.
Speaker Bio:Mohit Sharma is a PhD candidate at IIIT Delhi and an incoming postdoc at the Institute of Statistical Mathematics, Tokyo, and Tohoku University. His current work spans the broader area of responsible ML and approaches to deal with noisy, imperfect data. His broader interests include problems at the intersection of Economics and Machine learning, as well as several methods to brew coffee.
- 09 - 09 July,2026
- 12:00 pm - 01:00 pm
- Online