Cursory look suggests they quantize both the weights and functions during classification on the smart phone and during the forwards and backwards pass during training. But you still have to use a continuous version of your logic gate to get an error gradient during training, so I'm guessing they still do the training on "supercomputers" (a desktop with a gpu really).
Cursory look suggests they quantize both the weights and functions during classification on the smart phone and during the forwards and backwards pass during training. But you still have to use a continuous version of your logic gate to get an error gradient during training, so I'm guessing they still do the training on "supercomputers" (a desktop with a gpu really).