{"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/minima-possible-weights-a-homogenous-deep","title":"Minima Possible Weights: A Homogenous Deep Ensemble Method for Cross-Subject Motor Imagery Classification","arxiv_id":null,"date":"2025-02-18","proceeding":"IEEE Access 2025 2","authors":["Quang Pham Lam Dinh","Isao Nambu"],"abstract":"Motor Imagery (MI) systems in Brain-Computer Interface (BCI) research provide communication and control solutions for individuals with motor impairments, yet cross-subject classification remains\r\nchallenging due to substantial inter-subject variability. In this study, we propose the Minima Possible Weights\r\n(MPW) method, an unsupervised learning approach designed to enhance MI classification through ensemble\r\ndeep learning. MPW aggregates predicted probabilities from multiple models and selects the class with the\r\nlowest associated weight for the final prediction. We evaluated MPW against various ensemble learning\r\nand test-time adaptation methods using two benchmark datasets: BCI Competition IV Dataset 2a and\r\nPhysionetMI. Our results indicate that MPW achieves cross-subject classification accuracy of up to 64.75%\r\non BCI Competition IV Dataset 2a and 66.92% on PhysionetMI. Although the current performance is not\r\nyet sufficient for practical BCI applications, MPW shows potential in reducing calibration time and easing\r\nthe burden of adapting models to new subjects.","url_abs":"https://ieeexplore.ieee.org/document/10878971/metrics#metrics","url_pdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10878971","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":"minima-possible-weights-a-homogenous-deep","repo_url":"https://github.com/louiseblade/MPW_BCI_IV_2a","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"brain-computer-interface","task_name":"Brain Computer Interface"},{"task_slug":"ensemble-learning","task_name":"Ensemble Learning"},{"task_slug":"motor-imagery","task_name":"Motor Imagery"},{"task_slug":"test-time-adaptation","task_name":"Test-time Adaptation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}