This talk explores potential ways that large-scale pre-trained foundation models can significantly impact biomedical applications. Biological systems are extremely complex, and effective machine learning foundation models should leverage vast amounts of multimodal data to capture the underlying biomedical knowledge. We will demonstrate the integration of multiple modalities, including tabular data, 3D structures, sequences, text, images, etc. into various biomedical foundation models, aimed at solving a wide range of downstream tasks in few-shot and zero-shot settings. These models focus on different levels of biomedical applications, from molecules, proteins, to cells, tissues and electronic health records.

He is an assistant professor in the Department of Computer Science at Yale University. His research focus includes algorithms for graph neural networks, geometric embeddings, explainable models, and more recently, multi-modal foundation models involving relational reasoning. He is the author of many widely used GNN algorithms such as GraphSAGE, PinSAGE, and GNNExplainer. In addition, he has worked on a variety of applications of graph learning in physical simulations, social networks, knowledge graphs, neuroscience, and biotechnology. He developed the first billion-scale graph embedding services at Pinterest, and the graph-based anomaly detection algorithm at Amazon.

Artificial Intelligence today stands at a transformative crossroads where deep scientific inquiry meets urgent societal need. In this keynote, I offer a personal and institutional perspective on how AI can bridge molecular mechanisms and systems-level understanding to serve the public good. Drawing from my own experiences as an active AI researcher, educators, and cross-functional faculty-leader at a public university, I reflect on the evolution of AI as a research discipline and its expanding role in domains such as biology, education, and governance. As the inaugural Chief AI Officer of a public university, I will share with you stories of how we are building infrastructure, policy, and educational pathways to make AI accessible, responsible, and impactful for all students, faculty, and staff. From training the next generation of researchers and students, to deploying institution-wide platforms and shaping public-sector partnerships, this talk is a journey through proteins, policies, and purpose. It is admittedly biased by my own experiences but an invitation to share and learn from one another. In particular, I invite the audience to imagine their role not only as innovators and educators, but as stewards of AI in motion in the service of students and communities, expanding access and deepening the meaning of public impact of a public university in the age of artificially intelligent systems.

Dr. Amarda Shehu is Professor of Computer Science, Associate Dean for Research, and the inaugural Vice President and Chief AI Officer at George Mason University. She leads the university’s AI strategy across research, education, workforce development, and public engagement. She formerly served as Associate Vice President for Research, providing leadership for the Institute for Digital InnovAtion. She has launched multiple transdisciplinary centers, including the Center for AI Innovation and Economic Competitiveness, the Center for Infrastructure Security in the Era of AI, the Center for Cybersecurity Research, and the Provost Transdisciplinary Center for Advancing Human-Machine Partnerships. She is also the architect of Mason’s new M.S. in Artificial Intelligence degree program and chairs the university’s AI-in-Government Council, advancing AI collaboration across academia, industry, and public agencies. An active AI researcher, Dr. Shehu has published over 190 papers with students and collaborators. She is a Fellow of the American Institute for Medical and Biological Engineering, a member of the Virginia Academy of Sciences, Engineering, and Medicine, and a recipient of Virginia’s Outstanding Faculty Award from SCHEV, the NSF CAREER Award, and multiple university honors for research, teaching, and mentoring. Her research is supported by NSF, DoD, Air Force, SBA, and other agencies.