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

Yao Lu

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review


The outputs of a trained neural network contain much richer information than just a 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 zeroshot learning method achieves the state-of-the-art results on the ImageNet of over 20000 classes.
Original languageEnglish
Title of host publicationProceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence
Subtitle of host publicationNew York, New York, USA, 9–15 July 2016
EditorsSubbarao Kambhampati
ISBN (Electronic)978-1-57735-770-4
Publication statusPublished - 9 Jul 2016
MoE publication typeA4 Article in a conference publication
EventInternational Joint Conference on Artificial Intelligence - New York Hilton Midtown, New York, United States
Duration: 9 Jul 201615 Jul 2016
Conference number: 25

Publication series

NameInternational Joint Conferences on Artificial Intelligence
PublisherIAAA press
ISSN (Electronic)1045-0823


ConferenceInternational Joint Conference on Artificial Intelligence
Abbreviated titleIJCAI
CountryUnited States
CityNew York
Internet address

Fingerprint Dive into the research topics of 'Unsupervised Learning on Neural Network Outputs: with Application in Zero-shot Learning'. Together they form a unique fingerprint.

Cite this