Model-Compression / Lossless_Compression

We propose a lossless compression algorithm based on the NTK matrix for DNN. The compressed network yields asymptotically the same NTK as the original (dense and unquantized) network, with its weights and activations taking values only in {0, 1, -1} up to scaling.

https://proceedings.neurips.cc/paper_files/paper/2022/file/185087ea328b4f03ea8fd0c8aa96f747-Paper-Conference.pdf

Geek Repo

Github PK Tool