{"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/improving-generalization-via-scalable","title":"Improving Generalization via Scalable Neighborhood Component Analysis","arxiv_id":"1808.04699","date":"2018-08-14","proceeding":"ECCV 2018 9","authors":["Zhirong Wu","Alexei A. Efros","Stella X. Yu"],"abstract":"Current major approaches to visual recognition follow an end-to-end\nformulation that classifies an input image into one of the pre-determined set\nof semantic categories. Parametric softmax classifiers are a common choice for\nsuch a closed world with fixed categories, especially when big labeled data is\navailable during training. However, this becomes problematic for open-set\nscenarios where new categories are encountered with very few examples for\nlearning a generalizable parametric classifier. We adopt a non-parametric\napproach for visual recognition by optimizing feature embeddings instead of\nparametric classifiers. We use a deep neural network to learn the visual\nfeature that preserves the neighborhood structure in the semantic space, based\non the Neighborhood Component Analysis (NCA) criterion. Limited by its\ncomputational bottlenecks, we devise a mechanism to use augmented memory to\nscale NCA for large datasets and very deep networks. Our experiments deliver\nnot only remarkable performance on ImageNet classification for such a simple\nnon-parametric method, but most importantly a more generalizable feature\nrepresentation for sub-category discovery and few-shot recognition.","url_abs":"http://arxiv.org/abs/1808.04699v1","url_pdf":"http://arxiv.org/pdf/1808.04699v1.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":"improving-generalization-via-scalable","repo_url":"https://github.com/zhirongw/snca.pytorch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"improving-generalization-via-scalable","repo_url":"https://github.com/Microsoft/snca.pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[],"methods":[{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.04699","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}