{"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/multiple-instance-learning-a-survey-of","title":"Multiple Instance Learning: A Survey of Problem Characteristics and Applications","arxiv_id":"1612.03365","date":"2016-12-11","proceeding":null,"authors":["Marc-André Carbonneau","Veronika Cheplygina","Eric Granger","Ghyslain Gagnon"],"abstract":"Multiple instance learning (MIL) is a form of weakly supervised learning\nwhere training instances are arranged in sets, called bags, and a label is\nprovided for the entire bag. This formulation is gaining interest because it\nnaturally fits various problems and allows to leverage weakly labeled data.\nConsequently, it has been used in diverse application fields such as computer\nvision and document classification. However, learning from bags raises\nimportant challenges that are unique to MIL. This paper provides a\ncomprehensive survey of the characteristics which define and differentiate the\ntypes of MIL problems. Until now, these problem characteristics have not been\nformally identified and described. As a result, the variations in performance\nof MIL algorithms from one data set to another are difficult to explain. In\nthis paper, MIL problem characteristics are grouped into four broad categories:\nthe composition of the bags, the types of data distribution, the ambiguity of\ninstance labels, and the task to be performed. Methods specialized to address\neach category are reviewed. Then, the extent to which these characteristics\nmanifest themselves in key MIL application areas are described. Finally,\nexperiments are conducted to compare the performance of 16 state-of-the-art MIL\nmethods on selected problem characteristics. This paper provides insight on how\nthe problem characteristics affect MIL algorithms, recommendations for future\nbenchmarking and promising avenues for research.","url_abs":"http://arxiv.org/abs/1612.03365v1","url_pdf":"http://arxiv.org/pdf/1612.03365v1.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":"multiple-instance-learning-a-survey-of","repo_url":"https://github.com/macarbonneau/MILSurvey","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"multiple-instance-learning-a-survey-of","repo_url":"https://github.com/sswetank-CS/MIL","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"benchmarking","task_name":"Benchmarking"},{"task_slug":"document-classification","task_name":"Document Classification"},{"task_slug":"multiple-instance-learning","task_name":"Multiple Instance Learning"},{"task_slug":"survey","task_name":"Survey"},{"task_slug":"weakly-supervised-learning","task_name":"Weakly-supervised Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1612.03365","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}