Large Margin Deep Networks for Classification
This work addresses the challenge of enhancing robustness and generalization in deep learning models for classification tasks, representing an incremental improvement over existing margin methods.
The authors tackled the problem of applying large margin principles to deep networks by proposing a novel loss function that enforces margins at any chosen layers, demonstrating improved empirical results on MNIST, CIFAR-10, and ImageNet datasets for tasks like generalization from small training sets, corrupted labels, and adversarial robustness.
We present a formulation of deep learning that aims at producing a large margin classifier. The notion of margin, minimum distance to a decision boundary, has served as the foundation of several theoretically profound and empirically successful results for both classification and regression tasks. However, most large margin algorithms are applicable only to shallow models with a preset feature representation; and conventional margin methods for neural networks only enforce margin at the output layer. Such methods are therefore not well suited for deep networks. In this work, we propose a novel loss function to impose a margin on any chosen set of layers of a deep network (including input and hidden layers). Our formulation allows choosing any norm on the metric measuring the margin. We demonstrate that the decision boundary obtained by our loss has nice properties compared to standard classification loss functions. Specifically, we show improved empirical results on the MNIST, CIFAR-10 and ImageNet datasets on multiple tasks: generalization from small training sets, corrupted labels, and robustness against adversarial perturbations. The resulting loss is general and complementary to existing data augmentation (such as random/adversarial input transform) and regularization techniques (such as weight decay, dropout, and batch norm).