1 paper
Vladimir Protsenko, Mikhalina Kharkevich, Alexander Vashchilko +1
Neural network quantization aims to find a discrete representation of parameters that preserves the performance of a full-precision (FP) model as faithfully as possible. Enforcing…