Pruning from Scratch

作者：Yulong Wang, Xiaolu Zhang, Lingxi Xie, Jun Zhou, Hang Su, Bo Zhang, Xiaolin Hu · 发表于：AAAI Publications (The Association for the Advancement of Artificial Intelligence (AAAI)) · 年份：2020 · DOI：10.1609/aaai.v34i07.6910 · 被引用次数：143 · 研究领域：Domain Adaptation and Few-Shot Learning、Advanced Neural Network Applications、Anomaly Detection Techniques and Applications

Network pruning is an important research field aiming at reducing computational costs of neural networks. Conventional approaches follow a fixed paradigm which first trains a large and redundant network, and then determines which units (e.g., channels) are less important and thus can be removed. In this work, we find that pre-training an over-parameterized model is not necessary for obtaining the target pruned structure. In fact, a fully-trained over-parameterized model will reduce the search space for the pruned structure. We empirically show that more diverse pruned structures can be directly pruned from randomly initialized weights, including potential models with better performance. Therefore, we propose a novel network pruning pipeline which allows pruning from scratch with little training overhead. In the experiments for compressing classification models on CIFAR10 and ImageNet datasets, our approach not only greatly reduces the pre-training burden of traditional pruning methods, but also achieves similar or even higher accuracy under the same computation budgets. Our results facilitate the community to rethink the effectiveness of existing techniques used for network pruning.