About Me
Hi there and welcome to my website for the Distributed REsearch Apprenticeships for Master’s (DREAM) program! My name is Madison and I am a first-year Master’s student in Computer Science (CS) at Columbia University, with an expected graduation date of December 2023. I graduated from UCLA in 2021, where I obtained a B.S. in Cognitive Science and a B.A. in Human Biology and Society. Clearly, I gravitate towards interdisciplinary fields, and I typically describe my undergrad coursework as a blend of Computer Science, Psychology, Biology, and Sociology.
I am currently pursuing the Machine Learning track of Columbia’s MS in CS program. I am also a fellow of the Lustgarten Whitney Family Fellowship’s 2021 cohort. This program was designed to help support highly qualified students from underrepresented and nontraditional backgrounds as they pursue a career in computer science through Columbia University’s Bridge program. I hope to use my interdisciplinary approach throughout my academic and professional pursuits in the field of computer science, ultimately to create and advocate for technology that is prosocial, equitable, and empowering.
About My Advisor
Dr. Richard Zemel is a Professor of Computer Science at Columbia University’s Fu Foundation School of Engineering and Applied Science. His broad interests span the fields of machine learning, artificial intelligence, and cognitive neuroscience, and his recent research focuses on Robust ML, Algorithmic Fairness, Few-shot Learning, and Continual Learning.
Dr. Zemel’s research contributions include, but are not limited to, fairness and robustness in the context of representation learning, classification, and recommendation and ranking algorithms. He also developed the Toronto Paper Matching System and co-founded the Vector Institute of the University of Toronto.
Dr. Zemel received his BSc degree in History and Science from Harvard University in 1984 and his PhD in Computer Science from the University of Toronto in 1993. From 2000 up until his move to Columbia University, Dr. Zemel was a Professor of Computer Science at the University of Toronto—he is currently on leave from his positions at U of T and the Vector Institute. Prior to that, Dr. Zemel was an Assistant Professor in Computer Science and Psychology at the University of Arizona and a Postdoctoral Fellow at the Salk Institute and at Carnegie Mellon University.
See here for Dr. Zemel’s profiles at Columbia University and the University of Toronto’s Vector Institute.
About My Project
The research that I have conducted as part of the DREAM program builds on the research that I conducted alongside Jannik Wiedenhaupt in Dr. Zemel’s course at Columbia University, COMS 6998: Fair and Robust Algorithms (Fall 2022). Broadly speaking, this research project examines the implementation and ecosystem dynamics of recommender system (RS) models.
Recommender systems (RSs) are an increasingly powerful and ubiquitous tool in everyday life. A state of information-saturation has become the norm (in the western world), and RSs crucially help consumers navigate this information overload by reducing the amount of information that one is exposed to. However, in addition to developing new RS models, there is a growing body of research dedicated to examining RSs’ downstream effects, which have social, political, and economic implications. For example, past research has shown correlations between RSs and filter bubbles, homogenization, and (disparities in) fairness.
This research project consists of two key phases/objectives:
- Using the T-RECS simulation environment, long-term user behavior is estimated under various recommendation strategies, which include myopic, exploration-promoting, and fairness-constraining algorithms.
- A comparative analysis of these models is then conducted to analyze the interplay between accuracy, recommendation quality, homogeneity, and fairness within these RS ecosystems.
My Blog
My Blog (In progress)
My Contact Info
You can contact me via e-mail at madison.g.thantu@columbia.edu or at madisonthantu@gmail.com, and feel free to connect on LinkedIn.