{"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/learning-with-labels-of-existing-and","title":"A Multiclass Multiple Instance Learning Method with Exact Likelihood","arxiv_id":"1811.12346","date":"2018-11-29","proceeding":null,"authors":["Xi-Lin Li"],"abstract":"We study a multiclass multiple instance learning (MIL) problem where the\nlabels only suggest whether any instance of a class exists or does not exist in\na training sample or example. No further information, e.g., the number of\ninstances of each class, relative locations or orders of all instances in a\ntraining sample, is exploited. Such a weak supervision learning problem can be\nexactly solved by maximizing the model likelihood fitting given observations,\nand finds applications to tasks like multiple object detection and localization\nfor image understanding. We discuss its relationship to the classic\nclassification problem, the traditional MIL, and connectionist temporal\nclassification (CTC). We use image recognition as the example task to develop\nour method, although it is applicable to data with higher or lower dimensions\nwithout much modification. Experimental results show that our method can be\nused to learn all convolutional neural networks for solving real-world multiple\nobject detection and localization tasks with weak annotations, e.g.,\ntranscribing house number sequences from the Google street view imagery\ndataset.","url_abs":"http://arxiv.org/abs/1811.12346v2","url_pdf":"http://arxiv.org/pdf/1811.12346v2.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":"learning-with-labels-of-existing-and","repo_url":"https://github.com/lixilinx/MCMIL","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"multiple-instance-learning","task_name":"Multiple Instance Learning"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}