the archive
#neural network compression
2 resultsstat.ML2019
Group Pruning using a Bounded-Lp norm for Group Gating and Regularization
Chaithanya Kumar Mummadi, Tim Genewein, Dan Zhang +2
The paper proposes a gating mechanism and a bounded L1 regularizer to enable group-wise channel pruning in convolutional neural networks, achieving significant parameter reductions…
#model pruning#group sparsity#regularization#neural network compression
cs.CV2019
GDRQ: Group-based Distribution Reshaping for Quantization
Haibao Yu, Tuopu Wen, Guangliang Cheng +3
The paper introduces Scale-Clip, a distribution reshaping method that makes weights and activations more uniform-like, and a group-based quantization scheme that learns separate qu…
#low-bit quantization#distribution reshaping#group-based quantization#neural network compression