AI for High-Velocity Access to Complex, Big Data Structures

Participate via Teams webinar link

Using simulation to optimize chemical or metallurgical processes typically involves looking up physical or chemical data for hundreds of sets of input parameters. The relationship between input and output is often complex and non-linear, with no analytical function available. This work presents the development, implementation, and evaluation of a Deep Convolutional Neural Network (DCNN) for predicting four output parameters from five predictor parameters from a data set of this kind.

Presented by Xue-Cheng Tai and Ellen Nordgård-Hansen, both Chief Scientists at NORCE