{"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/optimizing-for-generalization-in-machine-1","title":"Optimizing for Generalization in Machine Learning with Cross-Validation Gradients","arxiv_id":"1805.07072","date":"2018-05-18","proceeding":null,"authors":["Shane Barratt","Rishi Sharma"],"abstract":"Cross-validation is the workhorse of modern applied statistics and machine\nlearning, as it provides a principled framework for selecting the model that\nmaximizes generalization performance. In this paper, we show that the\ncross-validation risk is differentiable with respect to the hyperparameters and\ntraining data for many common machine learning algorithms, including logistic\nregression, elastic-net regression, and support vector machines. Leveraging\nthis property of differentiability, we propose a cross-validation gradient\nmethod (CVGM) for hyperparameter optimization. Our method enables efficient\noptimization in high-dimensional hyperparameter spaces of the cross-validation\nrisk, the best surrogate of the true generalization ability of our learning\nalgorithm.","url_abs":"http://arxiv.org/abs/1805.07072v1","url_pdf":"http://arxiv.org/pdf/1805.07072v1.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":"optimizing-for-generalization-in-machine-1","repo_url":"https://github.com/sbarratt/crossval","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"hyperparameter-optimization","task_name":"Hyperparameter Optimization"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}