Abstract
Although deeper and larger neural networks have achieved better performance, the complex network structure and increasing computational cost cannot meet the demands of many resource-constrained applications. An effective way to address this problem is to make use of dynamic inference mechanism. Existing methods usually choose to execute or skip an entire specific layer, which can only alter the depth of the network. In this paper, we propose a novel method called Dynamic Multi-path Neural Network (DMNN), which provides more path selection choices in terms of network width and depth during inference. The inference path of the network is determined by a controller, which takes into account both historical and object category information. The proposed method can be easily incorporated into most modern network architectures. Experimental results on ImageNet and CIFAR-100 demonstrate the superiority of our method on both efficiency and overall classification accuracy. To be specific, we integrate DMNN into ResNet-101 and find that our method significantly outperforms its counterparts with an encouraging 45.1% FLOPs reduction.
Abstract (translated by Google)
URL
http://arxiv.org/abs/1902.10949