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.
24Updated last year

Alternatives and similar repositories for Lossless_Compression:

Users that are interested in Lossless_Compression are comparing it to the libraries listed below