Aggregated Residual Transformations for Deep Neural Networks

Citations Over TimeTop 1% of 2017 papers

Abstract

We present a simple, highly modularized network architecture for image classification. Our network is constructed by repeating a building block that aggregates a set of transformations with the same topology. Our simple design results in a homogeneous, multi-branch architecture that has only a few hyper-parameters to set. This strategy exposes a new dimension, which we call cardinality (the size of the set of transformations), as an essential factor in addition to the dimensions of depth and width. On the ImageNet-1K dataset, we empirically show that even under the restricted condition of maintaining complexity, increasing cardinality is able to improve classification accuracy. Moreover, increasing cardinality is more effective than going deeper or wider when we increase the capacity. Our models, named ResNeXt, are the foundations of our entry to the ILSVRC 2016 classification task in which we secured 2nd place. We further investigate ResNeXt on an ImageNet-5K set and the COCO detection set, also showing better results than its ResNet counterpart. The code and models are publicly available online.

Related Papers

• → Theories with models of prescribed cardinalities(1977)3 cited
• → A Note on Varying Cardinality in the Average Case Setting(1993)9 cited
• → Multi-Residual Networks: Improving the Speed and Accuracy of Residual Networks(2016)40 cited
• → On the cardinality of some powersets(1978)1 cited
• → Constructing sets of functions which have a givenF-cardinality(1992)