Slimmable Neural Networks (S-Net) is a novel network which enabled to select one of the predefined proportions of channels (sub-network) dynamically depending on the current computational resource availability. The accuracy of each sub-network on S-Net, however, is inferior to that of individually trained networks of the same size due to its difficulty of simultaneous optimization on different sub-networks. In this paper, we propose Slimmable Pruned Neural Networks (SP-Net), which has sub-network structures learned by pruning instead of adopting structures with the same proportion of channels in each layer (width multiplier) like S-Net, and we also propose new pruning procedures: multi-base pruning instead of one-shot or iterative pruning to realize high accuracy and huge training time saving. We also introduced slimmable channel sorting (scs) to achieve calculation as fast as S-Net and zero padding match (zpm) pruning to prune residual structure in more efficient way. SP-Net can be combined with any kind of channel pruning methods and does not require any complicated processing or time-consuming architecture search like NAS models. Compared with each sub-network of the same FLOPs on S-Net, SP-Net improves accuracy by 1.2-1.5% for ResNet-50, 0.9-4.4% for VGGNet, 1.3-2.7% for MobileNetV1, 1.4-3.1% for MobileNetV2 on ImageNet. Furthermore, our methods outperform other SOTA pruning methods and are on par with various NAS models according to our experimental results on ImageNet. The code is available at https://github.com/hideakikuratsu/SP-Net.
翻译:Slimmmable Neural 网络(S-SP-Net)是一个新颖的网络,它能够根据当前计算资源的可用性动态地选择频道(子网络)预设比例之一(次网络),但SNet上每个子网络的准确性不如受过个人训练的相同规模的网络,因为其在不同子网络上难以同时优化同步。在本文中,我们提议Slimmmable Promed Neural 网络(SP-Net),它拥有通过直线运行而学习的次网络结构,而不是采用像S-Net那样的每层(wids 倍倍)频道结构(wid-net),我们还提议新的运行程序:多基运行而不是一发或迭接线运行,以实现高精度和巨大的培训时间节约。我们还引入了较薄的频道排序(scs),以便以S-Net(zpm)的速度快速计算,以更有效率的方式运行,而不是采用S-Net-Net-maryal-Net(wise-Net) del-al-rmal-rmal-rmal-rma-rma-rma-rma-al-al-ral-al-rma-rma-s-ral-s-ral-s-s-ral-ral-s-s-s-s-s-s-s-s-s-s-s-al-al-ral-s-s-s-s-rm-rm-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-s-l-l-l-s-l-l-s-s-s-s-s-s-s-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l-l