Additive Powers-of-Two Quantization: An Efficient Non-uniform Discretization for Neural Networks.

Yuhang Li, Xin Dong, Wei Wang

Browse the full ICLR paper archive.