This is the official project repository for GACT: Activation Compressed Training for General Architectures by Xiaoxuan Liu, Lianmin Zheng, Dequan Wang, Yukuo Cen, Weize Chen, Xu Han, Jianfei Chen, ...
Genome-wide association studies (GWAS) have successfully identified genes associated with complex human diseases. Although much of the heritability remains unexplained, combining single nucleotide ...
GACT is an ACT framework that reduces memory footprint in training generic neural network architectures without prior domain knowledge, enabling larger batch sizes with negligible accuracy loss.
Some results have been hidden because they may be inaccessible to you
Show inaccessible results