The advancements of data-driven methods and machine learning (ML) have become pivotal in engineering, addressing crucial challenges such as material modeling and design, coupled-physics simulations, and inverse and data assimilation problems. The access to high-throughput experiments and digitized databases has enhanced data availability and is fueling an increased interest in data-driven techniques. These novel techniques are central to tackling complex challenges in modern solid mechanics applications. The rise of data-driven techniques encompasses physics-informed machine learning-based constitutive modeling and model-free approaches. By leveraging manifold learning and supervised/unsupervised reduced-order methods, these techniques have simplified linear and nonlinear, high-dimensional mechanics problems. This potential extends beyond modeling, offering accelerated material property predictions, enabling the discovery of new materials, and identifying the underlying mechanisms that govern complex material systems. Furthermore, integrating data-driven methods with computational mechanics has bridged the gap between theoretical predictions and real-world material behaviors, with applications ranging from optimizing structures to inverse design and solving partial differential equations (PDEs).

This mini-symposium aims to cover a broad spectrum of topics including, but not limited to:

Physics-informed machine learning-based constitutive modeling.
Model-free approaches in computational mechanics.
Supervised/Unsupervised reduced-order simulations of systems.
Causal discovery for interpretable modeling.
Data-assisted modeling and characterization of complex material systems.
Manifold and graph approaches for material modeling and design.
Machine learning-based optimization of structures.
Inverse design guided by generative machine learning.
Data-driven methods for solving partial differential equations (PDEs).