Cross-Task Inconsistency Based Active Learning (CTIAL) for Emotion Recognition

被引:1
作者
Xu, Yifan [1 ]
Jiang, Xue [1 ]
Wu, Dongrui [1 ]
机构
[1] Huazhong Univ Sci & Technol, Sch Artificial Intelligence & Automat, Key Lab, Minist Educ Image Proc & Intelligent Control, Wuhan 430074, Peoples R China
关键词
Task analysis; Uncertainty; Measurement uncertainty; Estimation; Entropy; Emotion recognition; Affective computing; Active learning; emotion classification; emotion estimation; transfer learning; FACIAL EXPRESSIONS;
D O I
10.1109/TAFFC.2024.3366767
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Emotion recognition is a critical component of affective computing. Training accurate machine learning models for emotion recognition typically requires a large amount of labeled data. Due to the subtleness and complexity of emotions, multiple evaluators are usually needed for each affective sample to obtain its ground-truth label, which is expensive. To save the labeling cost, this paper proposes an inconsistency-based active learning approach for cross-task transfer between emotion classification and estimation. Affective norms are utilized as prior knowledge to connect the label spaces of categorical and dimensional emotions. Then, the prediction inconsistency on the two tasks for the unlabeled samples is used to guide sample selection in active learning for the target task. Experiments on within-corpus and cross-corpus transfers demonstrated that cross-task inconsistency could be a very valuable metric in active learning. To our knowledge, this is the first work that utilizes prior knowledge on affective norms and data in a different task to facilitate active learning for a new task, even the two tasks are from different datasets.
引用
收藏
页码:1659 / 1668
页数:10
相关论文
共 40 条
[31]   Norms of valence, arousal, and dominance for 13,915 English lemmas [J].
Warriner, Amy Beth ;
Kuperman, Victor ;
Brysbaert, Marc .
BEHAVIOR RESEARCH METHODS, 2013, 45 (04) :1191-1207
[32]   Affective Brain-Computer Interfaces (aBCIs): A Tutorial [J].
Wu, Dongrui ;
Lu, Bao-Liang ;
Hu, Bin ;
Zeng, Zhigang .
PROCEEDINGS OF THE IEEE, 2023, 111 (10) :1314-1332
[33]   Transfer Learning for EEG-Based Brain-Computer Interfaces: A Review of Progress Made Since 2016 [J].
Wu, Dongrui ;
Xu, Yifan ;
Lu, Bao-Liang .
IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS, 2022, 14 (01) :4-19
[34]   Affect Estimation in 3D Space Using Multi-Task Active Learning for Regression [J].
Wu, Dongrui ;
Huang, Jian .
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING, 2022, 13 (01) :16-27
[35]   Active learning for regression using greedy sampling [J].
Wu, Dongrui ;
Lin, Chin-Teng ;
Huang, Jian .
INFORMATION SCIENCES, 2019, 474 :90-105
[36]   A Survey on Negative Transfer [J].
Zhang, Wen ;
Deng, Lingfei ;
Zhang, Lei ;
Wu, Dongrui .
IEEE-CAA JOURNAL OF AUTOMATICA SINICA, 2023, 10 (02) :305-329
[37]   Dynamic Active Learning Based on Agreement and Applied to Emotion Recognition in Spoken Interactions [J].
Zhang, Yue ;
Coutinho, Eduardo ;
Zhang, Zixing ;
Quan, Caijiao ;
Schuller, Bjorn .
ICMI'15: PROCEEDINGS OF THE 2015 ACM INTERNATIONAL CONFERENCE ON MULTIMODAL INTERACTION, 2015, :275-278
[38]   Speech emotion recognition based on hierarchical attributes using feature nets [J].
Zhao, Huijuan ;
Ye, Ning ;
Wang, Ruchuan .
INTERNATIONAL JOURNAL OF PARALLEL EMERGENT AND DISTRIBUTED SYSTEMS, 2020, 35 (03) :354-364
[39]  
Zhou F, 2020, NEUROCOMPUTING, V392, P38
[40]  
Zhou H, 2019, INT CONF ACOUST SPEE, P3732, DOI [10.1109/icassp.2019.8683299, 10.1109/ICASSP.2019.8683299]