Bezirganyan G. auteur aut IRD Sellami S. auteur aut IRD Berti-Equille Laure auteur aut IRD Fournier S. auteur aut IRD text bookSection eng text/pdf born digital access In this paper, we propose M2-Mixer, an MLP-Mixer based architecture with multi-head loss for multimodal classification. It achieves better performances than the convolutional, recurrent, or neural architecture search based baseline models with the main advantage of conceptual and computational simplicity. The proposed multi-head loss function addresses the problem of modality predominance (i.e., when one of the modalities is favored over the others by the training algorithm). Our experiments demonstrate that our multimodal mixer architecture, combined with the multi-head loss function, outperforms the baseline models on two benchmark multimodal datasets: AVMNIST and MIMIC-III with respectively, on average, + 0.43% in accuracy and 6. 4 times reduction in training time and + 0.33% in accuracy and 13. 3 times reduction in training time, compared with previous best performing models. specialized 122 He J. ed. edt IRD Palpanas T. ed. edt IRD Cuzzocrea A. ed. edt IRD Dou D. ed. edt IRD Slezak D. ed. edt IRD Wang W. ed. edt IRD Gruca A. ed. edt IRD Chun-Wei J. ed. edt IRD Agrawal R. ed. edt IRD 1052-1058 Piscataway IEEE 2023 IEEE International Conference on Big Data, Sorrento (ITA), 2023/12/15-2023/12/18 https://www.documentation.ird.fr/hor/fdi:010090983 10.1109/BigData59044.2023.10386252 979-8-3503-2446-4 [F B010090983] https://www.documentation.ird.fr/hor/fdi:010090983 https://www.documentation.ird.fr/intranet/publi/2024-08/010090983.pdf Accès réservé (Intranet de l'IRD) IRD - Base Horizon / Pleins textes 2024-06-28 2025-02-24 fdi:010090983 fre