{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/unsupervised-learning-on-neural-network","title":"Unsupervised Learning on Neural Network Outputs: with Application in Zero-shot Learning","arxiv_id":"1506.00990","date":"2015-06-02","proceeding":null,"authors":["Yao Lu"],"abstract":"The outputs of a trained neural network contain much richer information than\njust an one-hot classifier. For example, a neural network might give an image\nof a dog the probability of one in a million of being a cat but it is still\nmuch larger than the probability of being a car. To reveal the hidden structure\nin them, we apply two unsupervised learning algorithms, PCA and ICA, to the\noutputs of a deep Convolutional Neural Network trained on the ImageNet of 1000\nclasses. The PCA/ICA embedding of the object classes reveals their visual\nsimilarity and the PCA/ICA components can be interpreted as common visual\nfeatures shared by similar object classes. For an application, we proposed a\nnew zero-shot learning method, in which the visual features learned by PCA/ICA\nare employed. Our zero-shot learning method achieves the state-of-the-art\nresults on the ImageNet of over 20000 classes.","url_abs":"http://arxiv.org/abs/1506.00990v10","url_pdf":"http://arxiv.org/pdf/1506.00990v10.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"unsupervised-learning-on-neural-network","repo_url":"https://github.com/yaolubrain/ULNNO","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"}],"methods":[{"method_slug":"ica","method_name":"ICA"},{"method_slug":"pca","method_name":"PCA"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1506.00990","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}