Prototype-based sample-weighted distillation unified framework adapted to missing modality sentiment analysis

• Published in: Neural networks Vol. 177; p. 106397
• Main Authors: Zhang, Yujuan, Liu, Fang’ai, Zhuang, Xuqiang, Hou, Ying, Zhang, Yuling
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.09.2024
• Subjects: Knowledge distillation; Missing modality; Multimodal sentiment analysis; Optimization imbalance; Prototype network; Missing modality; Prototype network; Knowledge distillation; Multimodal sentiment analysis; Optimization imbalance
• ISSN: 0893-6080; 1879-2782; 1879-2782
• DOI: 10.1016/j.neunet.2024.106397

Missing modality sentiment analysis is a prevalent and challenging issue in real life. Furthermore, the heterogeneity of multimodality often leads to an imbalance in optimization when attempting to optimize the same objective across all modalities in multimodal networks. Previous works have consistently overlooked the optimization imbalance of the network in cases when modalities are absent. This paper presents a Prototype-Based Sample-Weighted Distillation Unified Framework Adapted to Missing Modality Sentiment Analysis (PSWD). Specifically, it fuses features with a more efficient transformer-based cross-modal hierarchical cyclic fusion module. Subsequently, we propose two strategies, namely sample-weighted distillation and prototype regularization network, to address the issues of missing modality and optimization imbalance. The sample-weighted distillation strategy assigns higher weights to samples that are located closer to class boundaries. This facilitates the obtaining of complete knowledge by the student network from the teacher’s network. The prototype regularization network calculates a balanced metric for each modality, which adaptively adjusts the gradient based on the prototype cross-entropy loss. Unlike conventional approaches, PSWD not only connects the sentiment analysis study in the missing modality to the full modality, but the proposed prototype regularization network is not reliant on the network structure and can be expanded to more multimodal studies. Massive experiments conducted on IEMOCAP and MSP-IMPROV show that our method achieves the best results compared to the latest baseline methods, which demonstrates its value for application in sentiment analysis. •This paper establishes a connection between the study of full and missing modality.•A sample-weighted distillation strategy is employed to adapt to missing modality.•A regularization network is proposed to mitigate the optimization imbalance.•Massive experimental results prove the superiority and robustness of our model.