Mechanism-Driven Al for Alloy Design: From Uncertainty-aware Physics-based ML Prediction of Plasticity to LLM-guided Symbolic Regression
Speaker: Dr. Jing Luo, Postdoctoral Researcher, Mechanical Engineering, Johns Hopkins University
Time: 23rd November, 2026, 3-4 pm (EST)
Location: Zoom Meeting ID: 815 9308 5776 | Passcode: 255409
Abstract
Designing alloys for extreme environments requires reliable predictions of mechanical behavior, yet materials data are often sparse and deformation mechanisms remain poorly understood. In this talk, I will present a mechanism-driven framework that integrates probabilistic machine learning, symbolic regression, and multiscale simulations across three areas:
- Physics-based, uncertainty-aware models that combine learned dislocation density evolution laws with strengthening theory to predict stress-strain responses from pure metals to multi-principal element alloys.
- An active-learning multi-agent system that automates feature and model selection, training, and experimental candidate recommendations, demonstrated on alloy spall strength and dislocation density.
- A self-evolving knowledge graph that organizes literature-derived processing-microstructure-property relationships through targeted searches to ensure traceable, physically consistent reasoning.
Together, these efforts lay the foundation for trustworthy, closed-loop alloy design where simulations, experiments, and knowledge infrastructure jointly guide mechanism discovery and material development.
About the Speaker
Dr. Jing Luo's research integrates artificial intelligence, multiscale simulations, and physical mechanisms to advance the design and manufacturing of alloys for extreme environments. Her work focuses on uncertainty-aware physics-based prediction of plasticity for alloys and knowledge-driven Al systems for alloy design.
Her long-term vision is to develop mechanism-driven, Al-enabled closed-loop systems accelerate materials discovery and design integrating simulations, experiments, and theory together.
Jing earned her PhD in Mechanical Engineering from Johns Hopkins University under the supervision of Prof. Jaafar El-Awady. She holds a master's degree in solid mechanics and bachelor's degrees in engineering mechanics and applied mathematics from Dalian University of Technology, China.
| Attachment | Size |
|---|---|
| Seminar-flyer-Jing-Luo.jpg | 453.66 KB |