Unsupervised Learning on Neural Network Outputs: with Application in Zero-shot Learning

Abstract

The outputs of a trained neural network contain much richer information than just an one-hot classifier. For example, a neural network might give an image of a dog the probability of one in a million of being a cat but it is still much larger than the probability of being a car. To reveal the hidden structure in them, we apply two unsupervised learning algorithms, PCA and ICA, to the outputs of a deep Convolutional Neural Network trained on the ImageNet of 1000 classes. The PCA/ICA embedding of the object classes reveals their visual similarity and the PCA/ICA components can be interpreted as common visual features shared by similar object classes. For an application, we proposed a new zero-shot learning method, in which the visual features learned by PCA/ICA are employed. Our zero-shot learning method achieves the state-of-the-art results on the ImageNet of over 20000 classes.

References in corpus (5)

• Distilling the Knowledge in a Neural Network
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Deep Neural Networks Rival the Representation of Primate IT Cortex for Core Visual Object Recognition
• Zero-Shot Learning Through Cross-Modal Transfer
• Distributional Smoothing with Virtual Adversarial Training

Cited by in corpus (10)

• Semantic Autoencoder for Zero-Shot Learning
• Synthesized Classifiers for Zero-Shot Learning
• Transductive Zero-Shot Learning with Visual Structure Constraint
• Visual Interpretability for Deep Learning: a Survey
• Zero-shot Recognition via Semantic Embeddings and Knowledge Graphs
• Transfer Learning for Speech and Language Processing
• Transductive Unbiased Embedding for Zero-Shot Learning
• Growing Interpretable Part Graphs on ConvNets via Multi-Shot Learning
• Zero and Few Shot Learning with Semantic Feature Synthesis and Competitive Learning
• A Novel Perspective to Zero-shot Learning: Towards an Alignment of Manifold Structures via Semantic Feature Expansion