Synthesis and Pruning as a Dynamic Compression Strategy for Efficient Deep Neural Networks

Open in new window