{"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/on-minimum-discrepancy-estimation-for-deep","title":"On Minimum Discrepancy Estimation for Deep Domain Adaptation","arxiv_id":"1901.00282","date":"2019-01-02","proceeding":null,"authors":["Mohammad Mahfujur Rahman","Clinton Fookes","Mahsa Baktashmotlagh","Sridha Sridharan"],"abstract":"In the presence of large sets of labeled data, Deep Learning (DL) has\naccomplished extraordinary triumphs in the avenue of computer vision,\nparticularly in object classification and recognition tasks. However, DL cannot\nalways perform well when the training and testing images come from different\ndistributions or in the presence of domain shift between training and testing\nimages. They also suffer in the absence of labeled input data. Domain\nadaptation (DA) methods have been proposed to make up the poor performance due\nto domain shift. In this paper, we present a new unsupervised deep domain\nadaptation method based on the alignment of second order statistics\n(covariances) as well as maximum mean discrepancy of the source and target data\nwith a two stream Convolutional Neural Network (CNN). We demonstrate the\nability of the proposed approach to achieve state-of the-art performance for\nimage classification on three benchmark domain adaptation datasets: Office-31\n[27], Office-Home [37] and Office-Caltech [8].","url_abs":"http://arxiv.org/abs/1901.00282v1","url_pdf":"http://arxiv.org/pdf/1901.00282v1.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":"on-minimum-discrepancy-estimation-for-deep","repo_url":"https://github.com/oezyurty/cluda","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1901.00282","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}