Somesh Jha
University of Wisconsin
Time: Friday, October. 24 12:30 PM - 1:30 PM Location: MKB 622
Zoom Link: https://tennessee.zoom.us/j/81554281291
Safety of AI through the lens of Security and Cryptography
Abstract:
AI techniques are being used in diverse domains where trustworthiness is a concern, including automotive systems, finance, healthcare, natural language processing, and malware detection. Of particular concern is the use of AI algorithms in cyber-physical systems (CPS), such as self-driving cars and aviation, where an adversary can cause serious consequences. Interest in this area of research has simply exploded. In this work, we will emphasize the need for a security and cryptography mindset in trustworthy machine learning, and then cover some lessons learned.
Bio:
Somesh Jha received his B.Tech from Indian Institute of Technology, New Delhi in Electrical Engineering. He received his Ph.D. in Computer Science from Carnegie Mellon University under the supervision of Prof. Edmund Clarke (a Turing award winner). Currently, Somesh Jha is the Lubar Professor in the Computer Sciences Department at the University of Wisconsin (Madison). His work focuses on analysis of security protocols, survivability analysis, intrusion detection, formal methods for security, and analyzing malicious code. Recently, he has focused on trustworthy ML. Somesh Jha has published several articles in highly-refereed conferences and prominent journals. He has won numerous best-paper and distinguished-paper awards. Prof. Jha received the CAV award for his work on CEGAR, and also has received the IIT-Delhi Distinguished Alumni award. Prof. Jha is the fellow of the ACM, IEEE, and AAAS.
Jenny Davis
Vanderbilt University
Time: Friday, Sept. 11 12:30 PM - 1:30 PM Location: MKB 622
After Algorithmic Fairness: The Myth of Neutrality and Power of Repair
Abstract:
Abstract: The field of algorithmic ethics is substantial and growing, working to mitigate harms and realize social good. The fairness paradigm dominates this field across AI, machine learning, and other data-driven domains. Algorithmic fairness aims to a) undercut human biases by replacing subjective assessments with 'objective' computation and b) eliminate biases in data and data-derived outputs. Despite significant investment from academia, industry, and government, algorithmic fairness has failed to live up to its promise. Algorithmic harms propagate and persist while social inequities amplify and embed. This talk presents algorithmic reparation as an alternate proposal. Drawing on a paper, collaborative workshop, special issue, and especially, aforthcoming book, the talk delineates a reparative paradigm for algorithmic futures. This begins with a critique of fairness as a viable value standard, making the case for a shift toward redress. This shift is supported by a tripartite framework of algorithmic reparation and its implementation across diverse uses-cases, along with careful consideration of the obstacles and inroads to reparative praxis.
Bio:
Jenny L. Davis is the Gertrude Conaway Vanderbilt Chair and Professor of Sociology at Vanderbilt University, Honorary Professor of Sociology at The Australian National University, and Non-Resident Fellow at the Center for Democracy and Technology.