Conference: NSF-NIH Joint Workshop on Foundational AI in Biology

Carnegie Mellon University, Pittsburgh PA

Investigators

Abstract

Methods and new techniques in artificial intelligence (AI) are rapidly being developed, extended, and applied to challenging problems in biology. At the same time, as new experimental methods, new data collection efforts, and greater understanding are developed in biology, the class and scope of problems that are amendable to AI approaches is growing. Extreme interest and rapid progress in AI has also made the AI research landscape diverse and complex, with a number of related, competing, and overlapping technologies and approaches being advanced. This rich ecosystem is a boon to research and innovation, but it also makes identifying threads and pressing problems more challenging. Thus, there is a need to survey the current frontier of the interface between AI methodology and biology and to chart future directions and challenges. We propose a 2-day, online workshop to address that need. This workshop will host invited talks by researchers working on foundational AI methods in biology. New directions and connections between emerging AI methodologies and problems in biology and health will be identified, synthesized, highlighted, and formalized. The workshop will catalyze discussion and help set the agenda for future research in this area, helping to launch new directions and solidify promising ones. The workshop will consist of approximately 13 talks and 4 discussion sessions over two days. Speakers will be selected based on their prior work and experience with developing new foundational AI methodology that answers biological questions. Speakers will be asked to focus on one or more of the following foundational AI themes as related to biology and human health: (1) Fairness and Social Effects of AI; (2) Federated Learning; (3) Generative Deep Learning Models; (4) Scalability of AI; (5) Privacy and Security in AI; (6) Method Optimization and Automated Algorithm Design; (7) Explainable AI and Causality; (8) Active Learning and Automated Science; (9) Transfer Learning; and (10) Incorporation of Prior Knowledge in AI. Though the workshop cannot hope to deeply cover all these foundational aspects, we aim to cover as many as possible. Speakers will connect these foundational computational techniques to problems in biology in areas such as genomics, structural biology, drug development, systems biology, biomedical imaging, neuroscience, and disease forecasting. With permission of the speaker, the talks will be recorded and distributed over the internet, and participants will be invited to contribute to a publication that surveys the identified themes, directions, and challenges. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.