Webinar - AlloyGPT: an agent-based LLM framework for the design of additively manufactured structural alloys in extreme environments

MACHINE LEARNING WEBINAR SERIES

AlloyGPT: an agent-based LLM framework for the design of additively manufactured structural alloys in extreme environments
Speaker: S. Mohadeseh Taheri-Mousavi
Tuesday January 21, 2025 at 2-3 PM CET/8-9 AM EST

Traditionally human are involved in various planning, collecting knowledge, designing, and validating structural alloys for a given application. This webinar will include a presentation of a large language model powered by collaborations of various AI agents to automate this process and accelerate the material discovery. The speaker will talk about how a framework was applied for the design of high-temperature strength printable Al alloys with thermal stability. As well as showcase how GPT-based LLMs which are trained by CALPHAD-based ICME data can predict microstructural features and properties in comparison with Bayesian optimization and conventional machine learning techniques.

Furthermore, there will be a discussion of the accuracy of the model in forward prediction and inverse design, its unique opportunities, and the efficiency in combining different agents in this design. This hybrid framework can lay the foundation for the automatic design of structural alloys which are manufactured by various techniques and are in extreme environments.

Register for the Webinar

Thermo-Calc Software’s Machine Learning Webinar Series

This webinar is a part of Thermo-Calc Software’s Machine Learning Webinar Series, taking place between November 2024 and February 2025. The series consists of five webinars. The first is by Thermo-Calc Software’s Research and Innovation group and showcases how we are using machine learning in our tools. The other four webinars feature Thermo-Calc users presenting work they have done using our tools together with machine learning. Find more information about the other webinars below.

Register for the Other Webinars

CALPHAD by Machine Learning and for Machine Learning │Wednesday November 20, 2024 at 2-3 PM CET/8-9 AM EST

Alloy Design Based on Artificial Intelligence and Machine Learning │Wednesday December 4, 2024 at 10-11 AM CET/4-5 PM CST

Bayesian Frameworks for Accelerated Alloy Discovery │Tuesday January 28, 2025 at 2-3 PM CET/8-9 AM EST

High-throughput CALPHAD calculations for screening and machine learning of refractory complex concentrated alloys │Tuesday February 4, 2024 at 2-3 PM CET/8-9 AM EST

About the Speaker

S. Mohadeseh Taheri-Mousavi joined Carnegie Mellon University as an assistant professor in September 2022 from MIT where she was a postdoctoral associate jointly in the Departments of Mechanical Engineering and Materials Science and Engineering. Before that, she was a postdoctoral fellow at Brown University. She received her Ph.D. from EPFL, Switzerland, and her B.Sc. and M.Sc. from Sharif University of Technology, Iran. She received both early and advanced prestigious Swiss National Science Foundation fellowships for her postdoctoral studies at Brown and MIT. The Taheri Mousavi Group develops novel multi-scale numerical and analytical frameworks in combination with machine learning techniques to discover next-generation structural alloys produced by various manufacturing techniques (particularly additive manufacturing) and under extreme environmental conditions. Their material informatics frameworks also guide experiments to be performed efficiently and in a smart manner. Taheri-Mousavi’s research was funded by large funding including grants from Naval Nuclear Laboratory, NASA STRI, DARPA, and Army Research Laboratory projects.

S. Mohadeseh Taheri-Mousavi
Assistant Professor
Departments of Materials Science and
Engineering and Mechanical Engineering,
Carnegie Mellon University