1 paper · 1 filter
Halyun Jeong, Jack Xin, Penghang Yin
Training quantized neural networks requires addressing the non-differentiable and discrete nature of the underlying optimization problem. To tackle this challenge, the straight-thr…