A key feature of neural networks, particularly deep convolutional neural networks, is their ability to "learn" useful representations from data. The very last layer of a neural network is then simply a linear model trained on these "learned" representations. Despite their numerous applications in other tasks such as classification, retrieval, clustering etc., a.k.a. transfer learning, not much work has been published that investigates the structure of these representations or indeed whether structure can be imposed on them during the training process. In this paper, we study the effective dimensionality of the learned representations by models that have proved highly successful for image classification. We focus on ResNet-18, ResNet-50 and VGG-19 and observe that when trained on CIFAR10, CIFAR100 and SVHN, the learned representations exhibit a fairly low rank structure. We propose a modification to the training procedure, which further induces low rank structure on learned activations. Empirically, we show that this has implications for robustness to adversarial examples and compression.
翻译:神经网络,特别是深层的进化神经网络的一个关键特征是它们能够从数据中“读取”有用的表达方式。神经网络的最后一层仅仅是一个关于这些“学到”的表述方式的线性模型。尽管它们在分类、检索、聚合等其它任务中应用了许多,例如转移学习,但是没有公布多少调查这些表达方式的结构或实际上在培训过程中是否可以强加给它们结构的工作。在本文件中,我们研究了已证明在图像分类方面非常成功的模型所学习的表述方式的有效维度。我们侧重于ResNet-18、ResNet-50和VGG-19,并观察到,在对CIFAR10、CIFAR100和SVHN等任务进行培训时,所学的表述方式显示出相当低的等级结构。我们建议修改培训程序,进一步为学习的激活活动带来低等级结构。我们很生动地表明,这对对抗性实例和压缩的强性有影响。