Research

I am particuarly interested in AI research positions in the finance, fintech, and government/regulatory space. Particular applications of interest include (but are not limited to) real-time payments and financial crime risk mitigation.

About Me

Hello! I am a final year PhD Candidate in computer science & engineering at the University of Michigan. I am advised by Michael Wellman as a member of the Strategic Reasoning Group.

My dissertation research lies at the intersection of computer science and economics. I apply artificial intelligence to study strategic decision making in financial networks. I gather data from simulation of agent-based models and use empirical game-theoretic analysis (EGTA) to reason about the decision making of various network participants. I aim to understand how strategic choices are made and the subsequent impacts of those decisions on the network. I have studied topics in debt cancellation, real-time payments, and financial fraud.

Prior to graduate school, I applied data analysis and machine learning methods to predict emergency room visits using cancer patient data and to predict events from social media data. As an undergraduate student, I attended the University of Massachusetts Amherst from which I obtained degrees in computer science and economics, as well as a minor in Chinese. My honors thesis, advised by Brendan O'Connor and Arindrajit Dube, applied time series forcasting to Google trends data for predicting state unemployment rates.

Research Interests

AI: computational economics, multi-agent systems, empirical game theory, applications of machine learning

Finance: real-time payments, fraud, systemic risk

Publications

Conference Publications

1. Fraud Risk Mitigation in Real-Time Payments: A Strategic Agent-Based Analysis
   Katherine Mayo, Nicholas Grabill, and Michael P. Wellman
   IJCAI '24: 33rd International Joint Conference on Artifical Intelligence, August 2024


2. A Strategic Analysis of Portfolio Compression
   Katherine Mayo and Michael P. Wellman
   ICAIF '21: 2nd ACM International Conference on AI in Finance, November 2021
   [paper][slides]
   Previous version presented as a poster at AAMAS 2021: [paper][poster]


3. An Agent-Based Model of Strategic Adoption of Real-Time Payments
   Katherine Mayo, Shaily Fozdar, Michael P. Wellman
   ICAIF '21: 2nd ACM International Conference on AI in Finance, November 2021
   [paper][slides]

Journal Publications

1. Machine Learning Model of Emergency Department Use for Patients Undergoing Treatment for Head and Neck Cancer Using Comprehensive Multifactor Electronic Health Records
   Michelle Mierzwa MD, Charles Mayo PhD, Pratyusha Yalamanchi MD, Joseph Evans MD, Francis Worden MD, Richard Medlin MD, Matt Schipper PhD, Caitlin Schonewolf MD, Jennifer Shah MD, Matthew Spector MD, Paul Swiecicki MD, Katherine Mayo MS, and Keith Casper MD
   JCO Clinical Cancer Informatics, January 2023
   [paper]

Workshop Papers

1. Flagging Payments for Fraud Detection: A Strategic Agent-Based Model
   Katherine Mayo, Shaily Fozdar, Michael P. Wellman
   MUFin '23: 3rd International Workshop on Modeling Uncertainty in the Financial World (@ 37th AAAI Conference on Artificial Intelligence), February 2023
   [paper][slides]

Coursework

Advanced Data Mining, Electronic Commerce (now called Incentives and Strategic Behavior in Computational Systems), International Finance, Reinforcement Learning Theory, Advanced Artificial Intelligence