{"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/hyperparameter-learning-for-conditional","title":"Hyperparameter Learning for Conditional Kernel Mean Embeddings with Rademacher Complexity Bounds","arxiv_id":"1809.00175","date":"2018-09-01","proceeding":null,"authors":["Kelvin Hsu","Richard Nock","Fabio Ramos"],"abstract":"Conditional kernel mean embeddings are nonparametric models that encode\nconditional expectations in a reproducing kernel Hilbert space. While they\nprovide a flexible and powerful framework for probabilistic inference, their\nperformance is highly dependent on the choice of kernel and regularization\nhyperparameters. Nevertheless, current hyperparameter tuning methods\npredominantly rely on expensive cross validation or heuristics that is not\noptimized for the inference task. For conditional kernel mean embeddings with\ncategorical targets and arbitrary inputs, we propose a hyperparameter learning\nframework based on Rademacher complexity bounds to prevent overfitting by\nbalancing data fit against model complexity. Our approach only requires batch\nupdates, allowing scalable kernel hyperparameter tuning without invoking kernel\napproximations. Experiments demonstrate that our learning framework outperforms\ncompeting methods, and can be further extended to incorporate and learn deep\nneural network weights to improve generalization.","url_abs":"http://arxiv.org/abs/1809.00175v3","url_pdf":"http://arxiv.org/pdf/1809.00175v3.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":"hyperparameter-learning-for-conditional","repo_url":"https://github.com/Kelvin-Hsu/cake","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}