{"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-boon-for-evaluating-architecture","title":"A Boo(n) for Evaluating Architecture Performance","arxiv_id":"1807.01961","date":"2018-07-05","proceeding":"ICML 2018 7","authors":["Ondrej Bajgar","Rudolf Kadlec","Jan Kleindienst"],"abstract":"We point out important problems with the common practice of using the best\nsingle model performance for comparing deep learning architectures, and we\npropose a method that corrects these flaws. Each time a model is trained, one\ngets a different result due to random factors in the training process, which\ninclude random parameter initialization and random data shuffling. Reporting\nthe best single model performance does not appropriately address this\nstochasticity. We propose a normalized expected best-out-of-$n$ performance\n($\\text{Boo}_n$) as a way to correct these problems.","url_abs":"http://arxiv.org/abs/1807.01961v2","url_pdf":"http://arxiv.org/pdf/1807.01961v2.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-boon-for-evaluating-architecture","repo_url":"https://gitlab.com/obajgar/boon","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","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}