{"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/on-the-importance-of-strong-baselines-in","title":"On the Importance of Strong Baselines in Bayesian Deep Learning","arxiv_id":"1811.09385","date":"2018-11-23","proceeding":null,"authors":["Jishnu Mukhoti","Pontus Stenetorp","Yarin Gal"],"abstract":"Like all sub-fields of machine learning Bayesian Deep Learning is driven by\nempirical validation of its theoretical proposals. Given the many aspects of an\nexperiment it is always possible that minor or even major experimental flaws\ncan slip by both authors and reviewers. One of the most popular experiments\nused to evaluate approximate inference techniques is the regression experiment\non UCI datasets. However, in this experiment, models which have been trained to\nconvergence have often been compared with baselines trained only for a fixed\nnumber of iterations. We find that a well-established baseline, Monte Carlo\ndropout, when evaluated under the same experimental settings shows significant\nimprovements. In fact, the baseline outperforms or performs competitively with\nmethods that claimed to be superior to the very same baseline method when they\nwere introduced. Hence, by exposing this flaw in experimental procedure, we\nhighlight the importance of using identical experimental setups to evaluate,\ncompare, and benchmark methods in Bayesian Deep Learning.","url_abs":"http://arxiv.org/abs/1811.09385v2","url_pdf":"http://arxiv.org/pdf/1811.09385v2.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":"on-the-importance-of-strong-baselines-in","repo_url":"https://github.com/yaringal/DropoutUncertaintyExps","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1811.09385","atlas_url":"https://app.syntology.ai/?focus=1811.09385","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}