Seth Watts | 20-ERD-020
Executive Summary
We will use machine learning techniques to develop accurate, efficient surrogate models for the complex response of micro-architected materials by training them on data sets we will generate using the finite element analysis of designs adaptively sampled from the full design space. The resulting surrogate models will have increased predictive power and reduced uncertainty relative to the current state of the art and can be implemented using existing design codes to support the nuclear weapons stockpile stewardship and national security missions.
