Abstract: Recent advances in machine learning have spawned innovation and prosperity in various fields. In machine learning models, nonlinearity facilitates more flexibility and ability to better fit the data. However, the improved model flexibility is often accompanied by challenges such as overfitting, less interpretability, and bias. Thus, my research has been focusing on designing new feasible nonlinear machine learning models to address the above different challenges, and bringing discoveries in both theory and applications. In this talk, I will introduce a new framework to improve the interpretability and fairness of deep learning models. From the model perspective, I will discuss our novel explainable and robust models, and provide rigorous theoretical guarantees on Pareto efficiency. From the data equity perspective, I will introduce our models to achieve fair predictions while being able to protect sensitive information. Last but not least, I will discuss the application of our framework in Alzheimer's disease study.

Bio: Xiaoqian Wang is an Assistant Professor of Electrical and Computer Engineering at Purdue University. She received her Ph.D. degree from the University of Pittsburgh in 2019, and the B.S. degree from Zhejiang University in 2013. She focuses on designing novel machine learning models to improve interpretability, fairness, and robustness. She also work on the intersection of machine learning and bioinformatics, healthcare.  She has published over 30 papers in top conferences and journals including NeurIPS, ICML, ICLR, AAAI, IJCAI, KDD, CVPR, RECOMB, IPMI, MICCAI, Bioinformatics, etc. She received an NSF CAREER award in 2022.

~BME Faculty Host: Dr. Young Kim ~

Zoom link: https://purdue-edu.zoom.us/j/7731446991?pwd=RHdkZTVnRkxTM3J3dnRvY1VLWTlYUT09

NOTE: Students registered for the seminar are expected to attend in-person.