{"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/knowledge-elicitation-via-sequential","title":"Knowledge Elicitation via Sequential Probabilistic Inference for High-Dimensional Prediction","arxiv_id":"1612.03328","date":"2016-12-10","proceeding":null,"authors":["Pedram Daee","Tomi Peltola","Marta Soare","Samuel Kaski"],"abstract":"Prediction in a small-sized sample with a large number of covariates, the\n\"small n, large p\" problem, is challenging. This setting is encountered in\nmultiple applications, such as precision medicine, where obtaining additional\nsamples can be extremely costly or even impossible, and extensive research\neffort has recently been dedicated to finding principled solutions for accurate\nprediction. However, a valuable source of additional information, domain\nexperts, has not yet been efficiently exploited. We formulate knowledge\nelicitation generally as a probabilistic inference process, where expert\nknowledge is sequentially queried to improve predictions. In the specific case\nof sparse linear regression, where we assume the expert has knowledge about the\nvalues of the regression coefficients or about the relevance of the features,\nwe propose an algorithm and computational approximation for fast and efficient\ninteraction, which sequentially identifies the most informative features on\nwhich to query expert knowledge. Evaluations of our method in experiments with\nsimulated and real users show improved prediction accuracy already with a small\neffort from the expert.","url_abs":"http://arxiv.org/abs/1612.03328v2","url_pdf":"http://arxiv.org/pdf/1612.03328v2.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":"knowledge-elicitation-via-sequential","repo_url":"https://github.com/HIIT/knowledge-elicitation-for-linear-regression","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"high","task_name":"Vocal Bursts Intensity Prediction"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1612.03328","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}