Connecting Data, Computing, and AI for Scientific Discovery
The post Connecting Data, Computing, and AI for Scientific Discovery appeared first on Berkeley Lab News Center .
Ana Kupresanin, Director of the Scientific Data Division at Lawrence Berkeley National Laboratory, explains how AI can revolutionize scientific discovery. Kupresanin, a statistician and Fellow of the American Statistical Association, leads efforts to develop methods, software, workflows, and infrastructure for making scientific data usable and reusable for science and AI.
She emphasizes that AI for science is about building systems grounded in scientific data, domain knowledge, uncertainty, and physical constraints rather than simply using the most advanced models. Berkeley Lab stands out in this endeavor due to its extensive expertise in scientific data, high-performance computing, simulations, AI models, and domain science.
The Lab's unique combination of facilities, computing capabilities, expertise in simulation and software, data management, statistics, uncertainty, and AI, along with domain experts, makes it well-positioned to contribute to the Department of Energy's Genesis Mission, a national initiative to advance AI for scientific discovery.
Kupresanin stresses the importance of scientific data in this context, noting that it carries context essential for trustworthy AI, such as how it was generated, assumptions, limitations, uncertainties, and reliable use cases.
Written by urgent.news from Berkeley Lab's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
