利用信息瓶颈的多模态情感分析

doi:10.3778/j.issn.1002-8331.2207-0456

摘要/Abstract

摘要： 在多模态情感分析领域，之前的研究主要集中在如何针对不同模态的信息进行交互融合。然而基于各种复杂的融合策略会使得生成的多模态表示向量不可避免地携带大量与下游任务无关的噪声信息，这会导致较高的过拟合风险，并且影响高质量预测结果的生成。为解决上述问题，根据信息瓶颈理论，设计了包含两个互信息估计器的互信息估计模块，旨在优化多模态表示向量与真实标签之间的互信息下界，同时最小化多模态表示向量与输入数据之间的互信息，以达到寻找一种简洁的、具有较好预测能力的多模态表示向量。利用MOSI、MOSEI和CH-SIMS数据集进行对比实验，结果表明提出的方法是富有成效的。

关键词: 多模态情感分析, 信息瓶颈理论, 互信息估计

Abstract: In the field of multimodal sentiment analysis, previous research mainly focused on how to interactively fuse information from different modalities. However, based on various complex fusion strategies, the generated multimodal representation vector inevitably carries a lot of noise information irrelevant to downstream tasks, which leads to a high risk of overfitting and affects the generation of high-quality prediction results. In order to solve the above problems, according to the information bottleneck theory, this paper designs a mutual information estimation module containing two mutual information estimators, aiming to optimize the lower bound of mutual information between the multimodal representation vector and the true label, while minimizing the multimodality. The mutual information between the representation vector and the input data is used to find a concise multimodal representation vector with better predictive ability. Using MOSI and MOSEI and CH-SMIS datasets to conduct comparative experiments, the results show that the method proposed in this paper is effective.

Key words: multimodal sentiment analysis, information bottleneck, mutual information estimating

程子晨, 李彦, 葛江炜, 纠梦菲, 张敬伟. 利用信息瓶颈的多模态情感分析[J]. 计算机工程与应用, 2024, 60(2): 137-146.

CHENG Zichen, LI Yan, GE Jiangwei, JIU Mengfei, ZHANG Jingwei. Multimodal Sentiment Analysis Based on Information Bottleneck[J]. Computer Engineering and Applications, 2024, 60(2): 137-146.

参考文献

[1] BALTRUSAITIS T, AHUJA C, MORENCY L P. Multimodal machine learning: a survey and taxonomy[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 41(2): 423-443.
[2] MAI S, HU H, XING S. A unimodal representation learning and recurrent decomposition fusion structure for utterance-level multimodal embedding learning[J]. IEEE Transactions on Multimedia, 2022, 24: 2488-2501.
[3] GRAVES A. Long short-term memory[M]//Supervised sequence labelling with recurrent neural networks. Berlin, Heidelberg: Springer, 2012, 385: 37-45.
[4] MAI S, XING S, HU H. Locally confined modality fusion network with a global perspective for multimodal human affective computing[J]. IEEE Transactions on Multimedia, 2020, 22(1): 122-137.
[5] MAI S, HU H, XING S. Divide, conquer and combine: hierarchical feature fusion network with local and global perspectives for multimodal affective computing[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Italy, 2019: 481-492.
[6] ZADEH A, CHEN M, PORIA S, et al. Tensor fusion network for multimodal sentiment analysis[C]//Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, Copenhagen, Denmark, 2017: 1103-1114.
[7] LIU Z, SHEN Y, LAKSHMINARASIMHAN V B, et al. Efficient low-rank multimodal fusion with modality-specific factors[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, Volume 1: (Long Papers), 2018: 2247-2256.
[8] 程子晨, 李彦, 葛江炜, 等. 利用跨模态调制的多模态情感分析[J]. 计算机工程与应用, 2023, 59(10): 171-179.
CHENG Z C, LI Y, GE J W, et al. Cross-modal modulating for multimodal sentiment analysis[J]. Computer Engineering and Applications, 2023, 59(10): 171-179.
[9] HAN W, CHEN H, PORIA S. Improving multimodal fusion with hierarchical mutual information maximization for multimodal sentiment analysis[C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, Dominican Republic, 2021: 9180-9192.
[10] ZENG Y, MAI S, HU H. Which is making the contribution: modulating unimodal and cross-modal dynamics for multimodal sentiment analysis[C]//Findings of the Association for Computational Linguistics: EMNLP 2021, Punta Cana, Dominican Republic, 2021: 1262-1274.
[11] MAI S, ZENG Y, ZHENG S, et al. Hybrid contrastive learning of tri-modal representation for multimodal sentiment analysis[J]. IEEE Transactions on Affective Computing, 2022, 14(3): 2276-2289.
[12] MAI S, HU H, XING S. Modality to modality translation: an adversarial representation learning and graph fusion network for multimodal fusion[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 164-172.
[13] YU W, XU H, YUAN Z, et al. Learning modality-specific representations with self-supervised multi-task learning for multimodal sentiment analysis[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2021: 10790-10797.
[14] WANG Y, WU J, FURUMAI K, et al. VAE-based adversarial multimodal domain transfer for video-level sentiment analysis[J]. IEEE Access, 2022, 10: 51315-51324.
[15] TISHBY N, PEREIRA F C, BIALEK W. The information bottleneck method[J]. arXiv:physics/0004057,2000.
[16] ALEMI A A, FISCHER I, DILLON J V, et al. Deep Variational information bottleneck[J]. arXiv:2310.03311,2023.
[17] ZAIDI A, ESTELLA-AGUERRI I, SHAMAI (SHITZ) S. On the information bottleneck problems: models, connections, applications and information theoretic views[J]. Entropy, 2020, 22(2): 151.
[18] WAN Z, ZHANG C, ZHU P, et al. Multi-view information-bottleneck representation learning[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2021: 10085-10092.
[19] MAI S, ZENG Y, HU H. Multimodal information bottleneck: learning minimal sufficient unimodal and multimodal representations[J]. IEEE Transactions on Multimedia, 2022: 4121-4134.
[20] BELGHAZI M I, BARATIN A, RAJESHWAR S, et al. Mutual information neural estimation[C]//Proceedings of the 35th International Conference on Machine Learning, 2018: 531-540.
[21] CHENG P, HAO W, DAI S, et al. CLUB: a contrastive log-ratio upper bound of mutual information[C]//Proceedings of the 37th International Conference on Machine Learning, 2020: 1779-1788.
[22] HAZARIKA D, ZIMMERMANN R, PORIA S. MISA: modality-invariant and -specific representations for multimodal sentiment analysis[C]//Proceedings of the 28th ACM International Conference on Multimedia. New York, NY, USA: Association for Computing Machinery, 2020: 1122-1131.
[23] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[C]//Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, December 4-9, 2017, Long Beach, CA, USA, 2017: 5998-6008.
[24] DEVLIN J, CHANG M W, LEE K, et al. BERT: pretraining of deep bidirectional transformers for language understanding[C]//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), 2019: 4171-4186.
[25] DELBROUCK J B, TITS N, DUPONT S. Modulated fusion using transformer for linguistic-acoustic emotion recognition[C]//Proceedings of the First International Workshop on Natural Language Processing Beyond Text, 2020.
[26] TSAI Y H H, BAI S, LIANG P P, et al. Multimodal transformer for unaligned multimodal language sequences[C]//Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. Florence, Italy: Association for Computational Linguistics, 2019: 6558-6569.
[27] CHENG J, FOSTIROPOULOS I, BOEHM B, et al. Multimodal phased transformer for sentiment analysis[C]//Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, 2021: 2447-2458.
[28] ZHANG Q, SHI L, LIU P, et al. ICDN: integrating consistency and difference networks by transformer for multimodal sentiment analysis[J]. Applied Intelligence, 2023, 53: 16332-16345.
[29] QI Q, LIN L, ZHANG R, et al. MEDT: using multimodal encoding-decoding network as in transformer for multimodal sentiment analysis[J]. IEEE Access, 2022, 10: 28750-28759.
[30] YANG B, SHAO B, WU L, et al. Multimodal sentiment analysis with unidirectional modality translation[J]. Neurocomputing, 2022, 467: 130-137.
[31] SUN H, WANG H, LIU J, et al. CubeMLP: an MLP-based model for multimodal sentiment analysis and depression estimation[C]//Proceedings of the 30th ACM International Conference on Multimedia, 2022: 3722-3729.
[32] RAHMAN W, HASAN M K, LEE S, et al. Integrating multimodal information in large pretrained transformers[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020.
[33] YANG K, XU H, GAO K. CM-BERT: cross-modal BERT for text-audio sentiment analysis[C]//Proceedings of the 28th ACM International Conference on Multimedia. New York, NY, USA: Association for Computing Machinery, 2020: 521-528.
[34] LUO H, JI L, HUANG Y, et al. ScaleVLAD: improving multimodal sentiment analysis via multi-scale fusion of locally descriptors[J]. arXiv:2112.01368,2021.
[35] TISHBY N, ZASLAVSKY N. Deep learning and the information bottleneck principle[C]//2015 IEEE Information Theory Workshop (ITW), 2015: 1-5.
[36] FEDERICI M, DUTTA A, FORRé P, et al. Learning robust representations via multi-view information bottleneck[J]. arXiv:2020.07017,2020.
[37] LEE C, SCHAAR M V D. A variational information bottleneck approach to multi-omics data integration[C]//Proceedings of the 24th International Conference on Artificial Intelligence and Statistics, 2021: 1513-1521.
[38] DONSKER M D, VARADHAN S R S. Asymptotic evaluation of certain Markov process expectations for large time—III[J]. Communications on Pure and Applied Mathematics, 1976, 29(4): 389-461.
[39] NGUYEN X, WAINWRIGHT M J, JORDAN M. Estimating divergence functionals and the likelihood ratio by penalized convex risk minimization[C]//Advances in Neural Information Processing Systems, 2007.
[40] NOWOZIN S, CSEKE B, TOMIOKA R. f-GAN: training generative neural samplers using variational divergence minimization[C]//Advances in Neural Information Processing Systems, 2016.
[41] ZADEH A, ZELLERS R, PINCUS E, et al. MOSI: multimodal corpus of sentiment intensity and subjectivity analysis in online opinion videos[J]. arXiv:1606.06259,2016.
[42] BAGHER ZADEH A, LIANG P P, PORIA S, et al. Multimodal language analysis in the wild: CMU-MOSEI dataset and interpretable dynamic fusion graph[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Melbourne, Australia: Association for Computational Linguistics, 2018: 2236-2246.
[43] YU W, XU H, MENG F, et al. CH-SIMS: a Chinese multimodal sentiment analysis dataset with fine-grained annotation of modality[C]//Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020: 3718-3727.
[44] DEGOTTEX G, KANE J, DRUGMAN T, et al. COVAREP—a collaborative voice analysis repository for speech technologies[C]//2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2014: 960-964.
[45] OORD A V D, LI Y, VINYALS O. Representation learning with contrastive predictive coding[J]. arXiv:1807.03748,2018.
[46] POOLE B, OZAIR S, VAN DEN OORD A, et al. On variational bounds of mutual information[C]//International Conference on Machine Learning, 2019: 5171-5180.