michaelhush/M-LOOP

Project details

M-LOOP currently using Gaussian process (GP) to model the experiment. The processing time required for a GP does not scale well with data (it's cubic). A solution to this scaling with large data issue is to use a neural net (NN). In this project we will add a NN to M-LOOP to do the learning instead of a GP. Or possibly in tandem with.