{"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/calibrated-boosting-forest","title":"Calibrated Boosting-Forest","arxiv_id":"1710.05476","date":"2017-10-16","proceeding":null,"authors":["Haozhen Wu"],"abstract":"Excellent ranking power along with well calibrated probability estimates are\nneeded in many classification tasks. In this paper, we introduce a technique,\nCalibrated Boosting-Forest that captures both. This novel technique is an\nensemble of gradient boosting machines that can support both continuous and\nbinary labels. While offering superior ranking power over any individual\nregression or classification model, Calibrated Boosting-Forest is able to\npreserve well calibrated posterior probabilities. Along with these benefits, we\nprovide an alternative to the tedious step of tuning gradient boosting\nmachines. We demonstrate that tuning Calibrated Boosting-Forest can be reduced\nto a simple hyper-parameter selection. We further establish that increasing\nthis hyper-parameter improves the ranking performance under a diminishing\nreturn. We examine the effectiveness of Calibrated Boosting-Forest on\nligand-based virtual screening where both continuous and binary labels are\navailable and compare the performance of Calibrated Boosting-Forest with\nlogistic regression, gradient boosting machine and deep learning. Calibrated\nBoosting-Forest achieved an approximately 48% improvement compared to a\nstate-of-art deep learning model. Moreover, it achieved around 95% improvement\non probability quality measurement compared to the best individual gradient\nboosting machine. Calibrated Boosting-Forest offers a benchmark demonstration\nthat in the field of ligand-based virtual screening, deep learning is not the\nuniversally dominant machine learning model and good calibrated probabilities\ncan better facilitate virtual screening process.","url_abs":"http://arxiv.org/abs/1710.05476v3","url_pdf":"http://arxiv.org/pdf/1710.05476v3.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":"calibrated-boosting-forest","repo_url":"https://github.com/haozhenWu/Calibrated-Boosting-Forest","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"},{"task_slug":"classification","task_name":"General Classification"},{"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}