{"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/transfer-learning-via-unsupervised-task","title":"Transfer Learning via Unsupervised Task Discovery for Visual Question Answering","arxiv_id":"1810.02358","date":"2018-10-03","proceeding":"CVPR 2019 6","authors":["Hyeonwoo Noh","Tae-hoon Kim","Jonghwan Mun","Bohyung Han"],"abstract":"We study how to leverage off-the-shelf visual and linguistic data to cope\nwith out-of-vocabulary answers in visual question answering task. Existing\nlarge-scale visual datasets with annotations such as image class labels,\nbounding boxes and region descriptions are good sources for learning rich and\ndiverse visual concepts. However, it is not straightforward how the visual\nconcepts can be captured and transferred to visual question answering models\ndue to missing link between question dependent answering models and visual data\nwithout question. We tackle this problem in two steps: 1) learning a task\nconditional visual classifier, which is capable of solving diverse\nquestion-specific visual recognition tasks, based on unsupervised task\ndiscovery and 2) transferring the task conditional visual classifier to visual\nquestion answering models. Specifically, we employ linguistic knowledge sources\nsuch as structured lexical database (e.g. WordNet) and visual descriptions for\nunsupervised task discovery, and transfer a learned task conditional visual\nclassifier as an answering unit in a visual question answering model. We\nempirically show that the proposed algorithm generalizes to out-of-vocabulary\nanswers successfully using the knowledge transferred from the visual dataset.","url_abs":"http://arxiv.org/abs/1810.02358v2","url_pdf":"http://arxiv.org/pdf/1810.02358v2.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":"transfer-learning-via-unsupervised-task","repo_url":"https://github.com/HyeonwooNoh/vqa_task_discovery","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"visual-question-answering-1","task_name":"Visual Question Answering"},{"task_slug":"visual-question-answering","task_name":"Visual Question Answering (VQA)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1810.02358","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}