Learning Discriminative Features via Label Consistent Neural Network
This addresses the issue of gradient vanishing and slow convergence in CNNs for researchers in computer vision, though it is incremental as it builds on existing neural network frameworks.
The authors tackled the problem of insufficient supervision in deep neural networks by proposing a Label Consistent Neural Network that enforces direct supervision in late hidden layers, resulting in state-of-the-art performances on action and object recognition benchmarks.
Deep Convolutional Neural Networks (CNN) enforces supervised information only at the output layer, and hidden layers are trained by back propagating the prediction error from the output layer without explicit supervision. We propose a supervised feature learning approach, Label Consistent Neural Network, which enforces direct supervision in late hidden layers. We associate each neuron in a hidden layer with a particular class label and encourage it to be activated for input signals from the same class. More specifically, we introduce a label consistency regularization called "discriminative representation error" loss for late hidden layers and combine it with classification error loss to build our overall objective function. This label consistency constraint alleviates the common problem of gradient vanishing and tends to faster convergence; it also makes the features derived from late hidden layers discriminative enough for classification even using a simple $k$-NN classifier, since input signals from the same class will have very similar representations. Experimental results demonstrate that our approach achieves state-of-the-art performances on several public benchmarks for action and object category recognition.