School of Computing Susan J. Rosowski Associate Professor Massimiliano Pierobon and Ph.D. student Shohaib Shaffiey co-authored a paper published at the 2026 International Conference on Machine Learning (ICML), held July 6–11 in Seoul, South Korea. Shaffiey attended and presented the paper at the conference.
The paper, “Tri-Scale Neural ODEs for Continuous Multi-Omics Disease Modeling,” explores how biological systems operate on different timescales, creating “stiff systems” that can make standard machine learning models difficult to train accurately and efficiently. The paper introduces a new architecture that mitigates this stiffness by separating the processes into distinct spatial and temporal scales.
“It tackles a major hurdle in AI-driven drug discovery,” Shaffiey said. “We developed a new machine learning architecture to handle the reality of how biological processes unfold over time. We ground our formulation with a mathematical framework, capturing everything from slow genetic changes to rapid metabolic responses. This gives us a much more comprehensive picture of disease progression and opens new avenues for drug repurposing.”
“Biological systems are inherently multiscale, with processes ranging from rapid molecular responses to much slower changes in gene regulation and disease progression,” Pierobon said. “Our goal is to develop machine learning methods that respect this structure rather than forcing biology into models that were not designed for it. Having this work published at ICML is particularly exciting because it brings these challenges at the intersection of machine learning, mathematics, and biology to one of the leading venues in the field.”
Shaffiey said he is grateful to have had the opportunity to present the paper at such a prestigious conference.
“Presenting in Seoul was one of the most rewarding experiences of my academic career,” Shaffiey said. “It was thrilling to discuss new solutions to current problems and explore potential collaborations with academic and industry leaders from places like Stanford, Columbia University, Yonsei University, Amazon, and Genentech.”
Pierobon said he is pleased with Shaffiey’s work as well as his presentation at the conference.
“I am very proud of Shohaib for presenting this work at ICML,” Pierobon said. “For a Ph.D. student, engaging directly with researchers from across academia and industry at a conference of this caliber is an invaluable opportunity, and it is wonderful to see his work receive that level of international exposure.”
The International Conference on Machine Learning (ICML) is the premier gathering of professionals dedicated to the advancement of the branch of artificial intelligence known as machine learning. ICML is globally renowned for presenting and publishing cutting-edge research on all aspects of machine learning used in closely related areas like artificial intelligence, statistics and data science, as well as important application areas such as machine vision, computational biology, speech recognition, and robotics.
