PhD Candidate
Purdue University
Areas of Interest
- Information, Data, Network, and Communication Sciences
Poster
Information Theoretical Modeling of Multi-Agent Interactions
Abstract
Addressing many major scientific and engineering challenges requires understanding and characterizing interactions in complex systems. For example, the risk of heart disease is associated with hundreds of genetic variants; global navigation systems require integrating information from multiple satellites; deep learning technologies depend on information processing through interconnected neurons. There is a great need for practical, yet theoretically grounded, methods for characterizing the low-level interactions that give rise to these various complex systems. The primary theme of my research is quantifying various types of complex interactions using tools from information theory and statistics. I proposed a novel differentiable measure of the complex interactions. I also found a novel duality between the information measures in communication channels with many-to-one and one-to-many network topologies. Building on this, my long-term goal is to develop rigorous methodologies for researchers in various domains to analyze and model complex systems.
Bio
Xueyan Niu received her B.S. degree in Mathematics and Applied Mathematics from Peking University, Beijing, China, in 2016. She is currently working toward the Ph.D. degree in industrial engineering at Purdue University, West Lafayette, IN, USA. Her research interests include information theory, multivariate statistical analysis, and applications including machine learning and computational neuroscience. She is a recipient of the ISITA2020 Outstanding Early Career Researcher Paper Award. Outside of research, she has a record of service through leadership roles in the Industrial Engineering Graduate Student Organization and the Industrial Engineering Graduate Women’s Group at Purdue.
