Neural Network Compression Via Sparse Optimization

Tianyi Chen, Bo Ji, Yixin Shi +4

The compression of deep neural networks (DNNs) to reduce inference cost becomes increasingly important to meet realistic deployment requirements of various applications. There have…