Hello!
I'm a final year (on the job market!) PhD student in Computer Science at TTIC, and I am fortunate to be advised by the marvelous Avrim Blum.
I work on making AI more reliable and robust, with the primary components being algorithmic fairness, strategic classification, and adversarial robustness.
I am particularly enthused about being able to work on trustworthy AI during this cambrian explosion of downstream tasks enabled by foundation models. Making models more robust and fair is critical and urgent and I intend to my leverage my skillset to help make that possible, using principled methods that play well with these big models in diverse contexts.
As a computer scientist, with a primarily technical skillset, my philosophy towards responsible AI is that my role is to push the pareto frontier forward and expose fundamental tradeoffs for stake-holders to evaluate, rather than me taking universally normative positions.
I did my undergraduate studies at UCLA in Applied Mathematics. My primary non-computer science academic interest is in American history. In my free time, I like drinking coffee, bicycling, and playing with my two cats, Figaro and Oleanna.
Talks and News
Two papers accepted to AISTATS 2024!
Interning at Intel Labs with Marius Arvinte and Cory Cornelius, figuring out if zero shot CLIP anomaly segmentation algorithms are robust to semantically consistent perturbations. January 2024-March 2024
Presenting our poster on Malicious Noise at the Neurips Fairness Workshop. December 2023
Interning at J.P. Morgan AI working with Margarita Boyarskaya on actionable counterfactuals using black box attacks. Working paper in progress. June 2023-Sept 2023
IDEAL: Workshop on Machine Learning, Interpretability, and Logic. Chicago, April 2023
UIC Computer Science Theory Seminar. Chicago (USA), March 2022
Papers
Bayesian Strategic Classification Lee Cohen, Saeed Sharifi-Malvajerdi, Kevin Stangl, Ali Vakilian, Juba Ziani, February 2024 (preprint)
Agnostic Multi-Robust Learning Using ERM Saba Ahmadi, Avrim Blum, Omar Montasser, and Kevin Stangl, AISTATS 2024
On the Vulnerability of Fairness Constrained Learning to Malicious Noise Avrim Blum, Princewill Okoroafor, Aadirupa Saha, Kevin Stangl, AISTATS 2024
Sequential Strategic Screening - Lee Cohen, Saeed Sharifi-Malvajerdi, Kevin Stangl, Ali Vakilian, Juba Ziani, ICML 2023
Multi Stage Screening: Enforcing Fairness and Maximizing Efficiency in a Pre-Existing Pipeline Avrim Blum, Kevin Stangl, Ali Vakilian, ACM FAccT 2022
Recovering From Biased Data: Can Fairness Constraints Improve Accuracy? - Avrim Blum and Kevin Stangl, FORC 2020 [talk]
Teaching
I TA'd TTIC 31020 Introduction to Machine Learning, taught by Kevin Gimpel in Fall 2019.
I was the TA for TTIC 31010 Algorithms, taught by Avrim Blum in Winter 2019, for which I received the TTIC Outstanding TA Award.