{"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/mondrian-forests-for-large-scale-regression","title":"Mondrian Forests for Large-Scale Regression when Uncertainty Matters","arxiv_id":"1506.03805","date":"2015-06-11","proceeding":null,"authors":["Balaji Lakshminarayanan","Daniel M. Roy","Yee Whye Teh"],"abstract":"Many real-world regression problems demand a measure of the uncertainty\nassociated with each prediction. Standard decision forests deliver efficient\nstate-of-the-art predictive performance, but high-quality uncertainty estimates\nare lacking. Gaussian processes (GPs) deliver uncertainty estimates, but\nscaling GPs to large-scale data sets comes at the cost of approximating the\nuncertainty estimates. We extend Mondrian forests, first proposed by\nLakshminarayanan et al. (2014) for classification problems, to the large-scale\nnon-parametric regression setting. Using a novel hierarchical Gaussian prior\nthat dovetails with the Mondrian forest framework, we obtain principled\nuncertainty estimates, while still retaining the computational advantages of\ndecision forests. Through a combination of illustrative examples, real-world\nlarge-scale datasets, and Bayesian optimization benchmarks, we demonstrate that\nMondrian forests outperform approximate GPs on large-scale regression tasks and\ndeliver better-calibrated uncertainty assessments than decision-forest-based\nmethods.","url_abs":"http://arxiv.org/abs/1506.03805v4","url_pdf":"http://arxiv.org/pdf/1506.03805v4.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":"mondrian-forests-for-large-scale-regression","repo_url":"https://github.com/harveydevereux/Mondrian","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"bayesian-optimization","task_name":"Bayesian Optimization"},{"task_slug":"gaussian-processes","task_name":"Gaussian Processes"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1506.03805","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}