AI & ML interests

Reduce neural network computing power requirements without sacrificing performance and accuracy.