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

arXiv: https://arxiv.org/abs/1506.00990 · HTML: https://mlsanity.com/paper/1506.00990/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Jun 2015.

## Built on by (most cited first, 4 of 4)

- [Exploiting a Joint Embedding Space for Generalized Zero-Shot Semantic Segmentation](https://mlsanity.com/paper/2108.06536/) (2021)
- [Exploring Hierarchical Graph Representation for Large-Scale Zero-Shot Image Classification](https://mlsanity.com/paper/2203.01386/) (2022)
- [Revisiting Document Representations for Large-Scale Zero-Shot Learning](https://mlsanity.com/paper/2104.10355/) (2021)
- [Semantic interpretation for convolutional neural networks: What makes a cat a cat?](https://mlsanity.com/paper/2204.07724/) (2022)

## Abstract (arXiv)

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.