The new CSAIL product amortizes the inference of your objects’ physics, indicating it trains a neural community to understand to predict the output of the normally costly statistical algorithm. Only a single pass through a neural network with minimal interaction knowledge is needed to simulate