Machine Learning in Mechanics and Materials

Organizers

Christos E. Athanasiou, Brown University
Nikolaos Bouklas, Cornell University
Steve Sun, Columbia University
Grace Gu, University of California Berkeley
Miguel Bessa, TU Delft

Description

The research field of engineering mechanics is becoming increasingly multidisciplinary resulting to problems of growing complexity especially in the fields of additive manufacturing, energy materials and bioengineering. Such complex problems that can be intrinsically high dimensional, have set the scene for the use of data-driven approaches (e.g., deep learning, Gaussian processes etc.) towards enhancing experimental as well as computational approaches. Indeed, in addition to the standard workflow of fitting a surrogate model to a large set of data to enable predictive capabilities, our community is seeking meaningful ways to integrate AI approaches both in data-rich and data-scarce regimes for a variety of problems. In this symposium, we aim to gather experts of the mechanics community, who work either on the development of state-of-the-art machine-learning enabled methods or on their employment in multidisciplinary mechanics problems. Some of the topics that will be discussed in our symposium are: Data-driven constitutive law discovery; AI and experimental mechanics; Machine learning in fracture mechanics; Data-driven manufacturing; Physics-informed machine learning.