{"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/automated-problem-identification-regression","title":"Automated Problem Identification: Regression vs Classification via Evolutionary Deep Networks","arxiv_id":"1707.00703","date":"2017-07-03","proceeding":null,"authors":["Emmanuel Dufourq","Bruce A. Bassett"],"abstract":"Regression or classification? This is perhaps the most basic question faced\nwhen tackling a new supervised learning problem. We present an Evolutionary\nDeep Learning (EDL) algorithm that automatically solves this by identifying the\nquestion type with high accuracy, along with a proposed deep architecture.\nTypically, a significant amount of human insight and preparation is required\nprior to executing machine learning algorithms. For example, when creating deep\nneural networks, the number of parameters must be selected in advance and\nfurthermore, a lot of these choices are made based upon pre-existing knowledge\nof the data such as the use of a categorical cross entropy loss function.\nHumans are able to study a dataset and decide whether it represents a\nclassification or a regression problem, and consequently make decisions which\nwill be applied to the execution of the neural network. We propose the\nAutomated Problem Identification (API) algorithm, which uses an evolutionary\nalgorithm interface to TensorFlow to manipulate a deep neural network to decide\nif a dataset represents a classification or a regression problem. We test API\non 16 different classification, regression and sentiment analysis datasets with\nup to 10,000 features and up to 17,000 unique target values. API achieves an\naverage accuracy of $96.3\\%$ in identifying the problem type without hardcoding\nany insights about the general characteristics of regression or classification\nproblems. For example, API successfully identifies classification problems even\nwith 1000 target values. Furthermore, the algorithm recommends which loss\nfunction to use and also recommends a neural network architecture. Our work is\ntherefore a step towards fully automated machine learning.","url_abs":"http://arxiv.org/abs/1707.00703v1","url_pdf":"http://arxiv.org/pdf/1707.00703v1.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":"automated-problem-identification-regression","repo_url":"https://github.com/yicli/API","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"},{"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}