Advancing Mechanics and Materials via Machine Learning

3-0-9 units · Graduate

Concepts in mechanics (solid mechanics: continuum, micro, meso, and molecular mechanics; elasticity, plasticity, fracture and buckling) and machine learning (stochastic optimization, neural networks, convolutional neural nets, adversarial neural nets, graph neural nets, recurrent neural networks and long/short-term memory nets, attention models, variational/autoencoders) introduced and applied to mechanics problems. Covers numerical methods, data and image processing, dataset generation, curation and collection, and experimental validation using additive manufacturing. Modules cover: foundations, fracture mechanics and size effects, molecular mechanics and applications to biomaterials (proteins), forward and inverse problems, mechanics of architected materials, and time dependent mechanical phenomena. Students taking graduate version complete additional assignments.