Artificial Intelligence for materials design: Advances and remaining challenges

Speaker: Adama Tandia, Science and Technology Division, Corning

Abstract
In the field of glass and polymer design, advanced analytical methods such as genetic algorithm, Gaussian processes, and neural networks can be applied to composition data with the resultant models used to design materials with highly specialized properties. The data used for such approaches can be empirical or generated using techniques such as molecular dynamics and density functional theory, among others. Yet, despite the many successes using these approaches, there remain serious challenges in terms of hands-free development of such accurate models and the robustness of the algorithms.

Bio
Dr. Adama Tandia received his Ph.D. in Applied Math/Applied Physics from Paul Sabatier University (France) in 1998. He immediately joined the department of Applied Mathematics at Northwestern University where he worked on application of Level Set methods on crystal growth. In 2000 Adama joined the department of Modeling & Simulation at Corning Incorporated. Adama has developed expertise in applications of molecular modeling for materials design, and development of machine learning algorithms for materials design and process optimization. Some of his current interests are Reinforcement Learning, Autonomous experimentation systems, Genetic Algorithm.