{"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/a-novel-bi-hemispheric-discrepancy-model-for-1","title":"A Novel Bi-hemispheric Discrepancy Model for EEG Emotion Recognition","arxiv_id":"1906.01704","date":"2019-05-11","proceeding":"arXiv:1906.01704 Search... Help | Advanced Search 2019 5","authors":[],"abstract":"The neuroscience study has revealed the discrepancy of emotion expression\nbetween left and right hemispheres of human brain. Inspired by this study, in\nthis paper, we propose a novel bi-hemispheric discrepancy model (BiHDM) to\nlearn the asymmetric differences between two hemispheres for\nelectroencephalograph (EEG) emotion recognition. Concretely, we first employ\nfour directed recurrent neural networks (RNNs) based on two spatial\norientations to traverse electrode signals on two separate brain regions, which\nenables the model to obtain the deep representations of all the EEG electrodes'\nsignals while keeping the intrinsic spatial dependence. Then we design a\npairwise subnetwork to capture the discrepancy information between two\nhemispheres and extract higher-level features for final classification.\nBesides, in order to reduce the domain shift between training and testing data,\nwe use a domain discriminator that adversarially induces the overall feature\nlearning module to generate emotion-related but domain-invariant feature, which\ncan further promote EEG emotion recognition. We conduct experiments on three\npublic EEG emotional datasets, and the experiments show that the new\nstate-of-the-art results can be achieved.","url_abs":"http://arxiv.org/abs/1906.01704v1","url_pdf":"http://arxiv.org/pdf/1906.01704v1.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":[],"tasks":[{"task_slug":"eeg-1","task_name":"EEG"},{"task_slug":"eeg-emotion-recognition","task_name":"EEG Emotion Recognition"},{"task_slug":"eeg","task_name":"Electroencephalogram (EEG)"},{"task_slug":"emotion-recognition","task_name":"Emotion Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/eeg-on-seed-iv","task":"Electroencephalogram (EEG)","dataset":"SEED-IV","model":"BiHDM","rank_in_archive_order":1,"of":3,"metrics":{"Accuracy":"74.35"},"uses_additional_data":false},{"leaderboard":"/sota/emotion-recognition-on-mped","task":"Emotion Recognition","dataset":"MPED","model":"BiHDM","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"40.34"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1906.01704","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}