I am a Ph.D. student at the Information, Risk and Operation Management Department at McCombs School of Business at the University of Texas at Austin. My research explores fairness and interpretability in socio-technical Systems. As part of my work, I characterize how bias may creep into data labels for training a supervised learning system, and how such label bias may lead to unwanted and undetected downstream consequences. Moreover, I explore methods that mitigate label bias and promote bias-aware active data collection. Generally speaking, in my research, I aim to understand the limits and risks of using machine learning (ML) systems and to develop methods to bring accountability to ML systems from both an algorithm design perspective and an ML designers’ behavioral perspective.
11/2022: Our paper “When More Data Lead Us Astray: Active Data Acquisition in the Presence of Label Bias” was accepted to HCOMP 2022, (Conference recording)
2021: Our work “Partially interpretable estimators (PIE): black-box-refined interpretable machine learning” is on Arxiv
2021: Our work “Gaussian process regression and classification using International Classification of Disease codes as covariates” is on Arxiv
2019: My summber intern project at UIHC AU Lab is published on Genome research