Abstract:
Artificial intelligence (AI) is transforming scientific discovery with autonomous laboratories emerging as a powerful paradigm for accelerating research and development. However, unlocking the full potential of autonomous experimentation requires AI systems that are not only efficient but also trustworthy, ensuring reliability, transparency, and robustness in decision-making. In the first part of this talk, Dr. Ewetz will present his recent research on enhancing trustworthiness in AI through explainable AI. This includes developing new methods and metrics for interpreting foundational AI models using feature attributions, path integrals, and information flow. Dr. Ewetz will then demonstrate how these techniques apply to robustness, retrieval-augmented generation, and knowledge editing. The second part of the talk will focus on the role of AI in automating programmable scientific laboratories. He will discuss digital twin development, natural language-to-temporal logic translation, and laboratory scheduling—leveraging neuro-symbolic AI, formal methods, and computer-aided design. Finally, Dr. Ewetz will conclude with insights into efficient AI and outline my future research vision.
Speaker’s Bio:
Dr. Rickard Ewetz is an associate professor in the Electrical and Computer Engineering (ECE) Department at the University of Florida. He received his Ph.D. in ECE from Purdue University in 2016. His research interests are broadly focused on the intersection of AI, hardware, and science, with a focus on explainability, neuro-symbolic AI, emerging hardware, and autonomous laboratories. Over the last ten years, he published over 90 peer-reviewed articles, including 22 publications through institutions on the prestigious Computer Science Rankings (CSRankings) list. Those include the Association for the Advancement of Artificial Intelligence (AAAI), the International Conference on Learning Representations (ICLR), the International Joint Conferences on Artificial Intelligence (IJCAI), the Design Automated Conference (DAC), the International Conference on Computer-Aided Design (ICCAD), and MicroAI's endpoint-based AI and machine learning engine (MICRO). His research has received four best paper nominations from top-tier venues such as ICCAD, the Design, Automation, and Test in Europe Conference (DATE, the Asia and South Pacific Design Automation Conference, and the Military Communications Conference. His research has been supported by the Defense Advanced Research Projects Agency, the Department of Energy, the National Science Foundation, the Air Force Research Laboratory, the Lockheed Martin Corporation, Cyber-Florida, and the Florida High Tech Corridor Council.
