Rose Yu: Towards AI Co-Scientists: Agentic AI for Scientific Discovery
About this Seminar
Towards AI Co-Scientists: Agentic AI for Scientific Discovery
Despite the huge success of agentic AI models across fields, they still suffer from hallucinations and can produce physically inconsistent outputs. Towards the grand dream of building AI co-scientists, it is critical to integrate physical laws, scientific simulations and formal methods into the learning and reasoning of these systems. In this talk, I will discuss our on-going effort to develop agentic AI models for physical sciences. I will demonstrate the use cases of our models on a variety of applications including earth science, epidemiology modeling, mathematics and physics.
https://bidmap.berkeley.edu/seminars/towards-ai-co-scientists-agentic-ai-scientific-discovery
Our Speaker:
Rose Yu
"Rose Yu is a professor at UC San Diego department of Computer Science and Engineering and Cofounder/CEO of GistFlow. She is a primary faculty with the AI Group. Her research interests lie primarily in machine learning, especially for large-scale spatiotemporal data. She is particularly excited about AI for scientific discovery. She has won Presidential Early Career Award for Scientists and Engineers (PECASE), DARPA Young Faculty Award, ECASE Award, NSF CAREER Award, Hellman Fellowship, Faculty Awards from Sony, JP Morgan, Meta, Google, Amazon, and Adobe, several Best Paper Awards, Best Dissertation Award at USC. She was named as MIT Technology Review Innovators Under 35 in AI and 2025 Samsung AI Researcher of the Year."
The Bakar Institute of Digital Materials for the Planet (BIDMaP) accelerates the discovery, development, and deployment of advanced materials to address some of the most urgent challenges facing our planet. By combining cutting-edge chemistry with artificial intelligence, machine learning, and robotics, BIDMaP is reimagining how materials can be designed and optimized for clean energy, clean air, clean water, advanced batteries, and sustainable chemical production.