{"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/a-simple-exponential-family-framework-for","title":"A Simple Exponential Family Framework for Zero-Shot Learning","arxiv_id":"1707.08040","date":"2017-07-25","proceeding":null,"authors":["Vinay Kumar Verma","Piyush Rai"],"abstract":"We present a simple generative framework for learning to predict previously\nunseen classes, based on estimating class-attribute-gated class-conditional\ndistributions. We model each class-conditional distribution as an exponential\nfamily distribution and the parameters of the distribution of each seen/unseen\nclass are defined as functions of the respective observed class attributes.\nThese functions can be learned using only the seen class data and can be used\nto predict the parameters of the class-conditional distribution of each unseen\nclass. Unlike most existing methods for zero-shot learning that represent\nclasses as fixed embeddings in some vector space, our generative model\nnaturally represents each class as a probability distribution. It is simple to\nimplement and also allows leveraging additional unlabeled data from unseen\nclasses to improve the estimates of their class-conditional distributions using\ntransductive/semi-supervised learning. Moreover, it extends seamlessly to\nfew-shot learning by easily updating these distributions when provided with a\nsmall number of additional labelled examples from unseen classes. Through a\ncomprehensive set of experiments on several benchmark data sets, we demonstrate\nthe efficacy of our framework.","url_abs":"http://arxiv.org/abs/1707.08040v3","url_pdf":"http://arxiv.org/pdf/1707.08040v3.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":"a-simple-exponential-family-framework-for","repo_url":"https://github.com/vkverma01/Zero-Shot","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"a-simple-exponential-family-framework-for","repo_url":"https://github.com/vkverma01/Zero-Shot-Learning","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"few-shot-learning","task_name":"Few-Shot Learning"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.08040","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}