Mathematics Research Communities

MRC Conference Week 2: June 23-29, 2024

Mathematics of Adversarial, Interpretable, and Explainable AI


  • Karamatou Yacoubou Djima, Wellesley College
  • Tegan Emerson, Pacific Northwest National Laboratory; Colorado State University; University of Texas El Paso
  • Emily King, Colorado State University
  • Dustin Mixon, The Ohio State University
  • Tom Needham, Florida State University

Some of the most active areas of research in machine learning today are adversarial artificial intelligence (AI), explainable AI, and interpretable AI. Most progress in these areas has been empirical and rooted in computer science, but there is a growing body of literature that suggests that fresh insights are available in fields that are traditionally considered to be pure mathematics, such as algebra, geometry, topology, and analysis. The goal of this MRC is to introduce early career mathematicians, coming from a variety of subdisciplines, to these cutting-edge research areas and to show them how they can use their own expertise to make substantial contributions.

In explainable AI, methods are developed “open up” the black boxes like neural networks, while interpretable AI creates white box methods with possibly lower accuracy. It is possible to “trick” a trained neural network into outputting an error; adversarial AI is the study of such phenomena, which could yield dangerous outcomes.

In addition to research breakout groups, this MRC will include several complementary activities like lectures on algebra, geometry, topology, and analysis in data science and their interaction with adversarial, interpretable, and explainable AI and career-focused and soft-skills training opportunities.

Applications will be accepted on through Thursday, February 15, 2024 (11:59 p.m. EST).