Monday  |  Salon 12  |  09:40 AM–10:20 AM

#16037, An Introduction to Machine Learning for Mechanics

Advances in metrology and computational power have launched the “big data” age, where conventional analysis tools cannot reasonably utilize all available data. Machine learning (ML) can be a disruptive capability that reframes how we approach our data. This talk is a short introduction to ML, meant to support the novice to appreciate the application of ML to mechanics. Topics include (1) an explanation of the primary goals and tasks of ML including classification, anomaly detection, and regression; (2) common methods such as neural networks and genetic algorithms; and (3) introduction to terminology including loss function and activation function. A few examples of ML applications illustrate how a mechanican can effectively learn from and evaluate the use of ML in mechanics. SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525.

Sharlotte Kramer   Sandia National Laboratories