Very Deep Convolutional Networks for Large-Scale Image Recognition
Citations
529 citations
Cites background or methods from "Very Deep Convolutional Networks fo..."
...VGG11* on CIFAR: We train a modified version of the popular 11-layer VGG11 network [28] on the CIFAR [30] data set....
[...]
...We run preliminary experiments with a simplified version of the well-studied 11-layer VGG11 network [28], which we train on the CIFAR-10 [30] data set in a federated learning setup using ten clients....
[...]
...0We denote by VGG11* a simplified version of the original VGG11 architecture described in [28], where all dropout and batch normalization layers are removed and the number of convolutional filters and size of all fully connected layers is reduced by a factor of 2....
[...]
528 citations
527 citations
527 citations
Cites background or methods from "Very Deep Convolutional Networks fo..."
...We study the performance of different algorithms on the mainstream CNN models, including VGGNet with a plain structure [32], GoogLeNet with an inception module [33], ResNet with a residual block [11] and DenseNet with a dense block [15]....
[...]
...An illustration of mainstream network structures to be pruned, including Plain structure [32], Inception module [33], Residual block [11] and Dense block [15]....
[...]
...We conduct extensive experiments on two benchmarks, CIFAR-10 [17] and ImageNet [31], using many representative large CNN models, including VGGNet [32], GoogLeNet [33], ResNet [11] and DenseNet [15]....
[...]
525 citations
References
49,639 citations
21,729 citations
9,803 citations
9,775 citations
6,061 citations