Network Pruning

Network pruning is widely used for reducing the inference cost of deep models for resource limited applications. The key research problems here include: which weights are to be pruned, when to prune and how to induce sparsity for better pruning.

Highlighted Work:

Learning Efficient Convolutional Networks through Network Slimming

Zhuang Liu, Jianguo Li, Zhiqiang Shen, Gao Huang, Shoumeng Yan, Changshui Zhang

ICCV 2017   ·   23 Aug 2017   ·   arxiv:1708.06519

Paper

Code