本研究的目的是利用情緒辨識技術結合不同的去中心化聯合學習架構來偵測人們的情緒狀態。情緒辨識技術是人工智慧領域的一個重要應用，它可以根據人類臉部肌肉的變化來確定一個人的情緒狀態。這些技術可以應用在各種場景中，例如心理健康評估、人機互動、教育和市場行銷等。而聯合學習則是一種分散式機器學習方法，允許多個客戶端一起學習並提高整體模型的準確性，從而無需將位處本地端的隱私性資料傳輸到中央伺服器。這不僅提升了資料安全性，還可以有效地處理大規模資料，同時減少網絡帶寬的需求。

儘管過去的情感辨識研究已經取得了一些顯著的成果，但情感辨識技術在非面對面交流中的應用仍存在許多挑戰。隨著資料隱私問題日益受到重視，如何有效蒐集和分析情感數據，同時保護個人隱私成為一個大問題。傳統的集中式學習方法需要將大量的數據上傳到中央伺服器，這可能導致個人隱私的洩露。在大量客戶端分散的情況下，聯合學習通過僅共享模型參數而非原始數據，有效地減少了隱私洩露的風險，並且可以只蒐集必要的訓練資料以提高模型精確度。

本研究的核心是在去中心化的聯合學習環境中，結合情緒辨識技術來探討不同的架構和演算法如何影響模型的性能。去中心化聯合學習是一種新興的技術，它不依賴於中央伺服器，而是允許多個設備（如手機、電腦等）在本地端進行模型訓練，然後將訓練所得的模型參數共享至全局，從而在保護隱私的前提下實現高效的分散式學習。這種方法特別適合於需要保護敏感數據的應用場景，如醫療健康、金融等領域。

在本文中，我們主要分析情緒辨識技術及其在各種不同情況下的表現。首先，我們回顧了現有的情緒辨識技術，包括傳統的機器學習方法和基於深度學習的方法。傳統的情緒辨識方法主要依賴於特徵工程，如面部肌肉的運動分析、眼睛運動的追蹤等，而深度學習方法則通過自動學習數據的特徵來進行分類。隨著深度學習技術的進步，特別是卷積神經網絡（CNN）的廣泛應用，情緒辨識的準確性和實時性都有了顯著提高。

接著，我們探討了聯合學習在情緒辨識中的應用，並設計了一個能夠在多個分散式客戶端之間共享和協作學習的架構來實現聯合學習系統。我們選擇了幾種不同的聯合學習演算法，包括：FedAvg、SCAFFOLD演算法，並在不同的實驗設置下進行了比較分析。我們的實驗結果顯示，聯合學習不僅可以顯著提升情緒辨識模型的準確性，還可以有效地解決數據異質性和系統異質性問題，這在實際應用中具有重要意義。

本研究結果證明，面部情緒識別加上聯合學習演算法可以顯著提高情緒辨識模型的精準度。實驗結果顯示，使用恰當的模型架構與對應的聯合學習演算法，最佳可以獲得 85% 的準確度。這一結果表明，通過合理設計聯合學習的架構和選擇合適的演算法，情緒辨識系統可以在保護隱私的同時達到較高的準確性。此外，我們的研究還揭示了不同的聯合學習策略對模型性能的影響，為未來的研究提供了有價值的參考。

因此，我相信這項研究將推動情感識別技術的發展，為提高人們的生活品質和工作效率發揮積極作用。隨著技術的進一步成熟，未來情緒辨識技術結合聯合學習有望在更多的應用場景中發揮作用，如智能客服、虛擬助理、在線教育等，這些應用將能夠更好地理解和回應使用者的情緒需求，提升人機交互的體驗。同時，去中心化聯合學習架構的應用也將為其他領域的研究提供新的思路，如醫療診斷、智能監控等，從而為社會帶來更廣泛的影響。

This study investigates using emotion recognition technology combined with decentralized federated learning to detect emotional states. Emotion recognition identifies emotions based on facial muscle changes and is applied in mental health, human-computer interaction, and marketing. Federated learning enables multiple clients to improve model accuracy without sharing private data, enhancing security and managing large-scale data efficiently.

Challenges remain in emotion recognition, particularly concerning privacy in non-face-to-face interactions. Federated learning addresses these concerns by sharing only model parameters, reducing privacy risks while improving accuracy.

This study examines how different architectures and algorithms impact performance in a decentralized federated learning environment. It reviews traditional and deep learning-based methods like CNNs, showing significant accuracy improvements. A federated learning system was designed and tested with algorithms like FedAvg and SCAFFOLD, revealing that federated learning enhances accuracy and addresses data and system heterogeneity issues.

Results show that combining emotion recognition with federated learning achieves up to 85% accuracy, demonstrating the potential for high accuracy while protecting privacy. The study concludes that this approach will advance emotion recognition technology, with applications in intelligent customer service and online education, improving human-computer interaction while safeguarding privacy.

[1] K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770-778, 2016.
[2] H. B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, "Communication-efficient learning of deep networks from decentralized data," in Proceedings of the 20th International Conference on Artificial Intelligence and Statistics (AISTATS), pp. 1273-1282, 2017.
[3] Womany, "Six basic emotions diagram." https://liff.line.me/1454987169-1WAXAP3K/v2/article/YaeKl8P?utm_source=copyshare, 2023.
[4] Kaggle, "Face expression recognition dataset." https://www.kaggle.com/datasets/jonathanoheix/face-expression-recognition-dataset, 2018.
[5] M. Tan and Q. V. Le, "Efficientnet: Rethinking model scaling for convolutional neural networks," in Proceedings of the 36th International Conference on Machine Learning (ICML), pp. 6105-6114, 2019.
[6] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, vol. 521, no. 7553, pp. 436-444, 2015.
[7] C. N. Anagnostopoulos, I. E. Anagnostopoulos, I. D. Psoroulas, V. Loumos, and E. Kayafas, "License plate recognition from still images and video sequences: A survey," IEEE Transactions on Intelligent Transportation Systems, vol. 9, no. 3, pp. 377-391, 2008.
[8] A. Kumar, A. Kumar, and K. Nguyen, "Liveness detection for face recognition in mobile applications," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 1-9, 2018.
[9] W. Wang, D. Yang, Y. Li, Y. Ma, Y. Song, J. Luo, and Y. Yin, "Facial expression recognition based on local binary patterns and kernel discriminant isomap," Sensors, vol. 16, no. 7, p. 1132, 2016.
[10] P. Ekman and W. V. Friesen, "Constants across cultures in the face and emotion," Journal of Personality and Social Psychology, vol. 17, no. 2, pp. 124-129, 1971.
[11] Z. Zeng, M. Pantic, G. I. Roisman, and T. S. Huang, "A survey of affect recognition methods: Audio, visual, and spontaneous expressions," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 31, no. 1, pp. 39-58, 2009.
[12] A. G. Howard, M. Zhu, B. Chen, D. Kalenichenko, W. Wang, T. Weyand, et al., "Mobilenets: Efficient convolutional neural networks for mobile vision applications," arXiv preprint arXiv:1704.04861, 2017.
[13] S. P. Karimireddy, S. Kale, M. Mohri, S. J. Reddi, S. U. Stich, and A. T. Suresh, "Scaffold: Stochastic controlled averaging for federated learning," arXiv preprint arXiv:1910.06378, 2019.
[14] M. Chawla et al., "A lightweight and privacy-preserving federated learning system for verbal communication analysis," Journal of Machine Learning Research, vol. 24, no. 4, pp. 123-145, 2024.
[15] S. Gahlan and M. Sethia, "Attention-based federated learning for emotion recognition using multimodal physiological data," IEEE Transactions on Affective Computing, vol. 15, no. 2, pp. 67-79, 2024.
[16] N. Rieke, J. Hancox, W. Li, F. Milletari, H. R. Roth, S. Albarqouni, et al., "The future of digital health with federated learning," NPJ Digital Medicine, vol. 3, no. 1, pp. 1-7, 2020.
[17] S. Savazzi, M. Nicoli, M. Bennis, and M. Debbah, "Opportunities of federated learning in connected, cooperative, and automated industrial systems," IEEE Communications Magazine, vol. 59, no. 2, pp. 16-21, 2020.
[18] P. Kairouz, H. B. McMahan, B. Avent, A. Bellet, M. Bennis, A. N. Bhagoji, et al., "Advances and open problems in federated learning," arXiv preprint arXiv:1912.04977, 2019.
[19] T. Li, A. K. Sahu, A. Talwalkar, and V. Smith, "Federated learning: Challenges, methods, and future directions," IEEE Signal Processing Magazine, vol. 37, no. 3, pp. 50-60, 2020.
[20] A. Hard, K. Rao, R. Mathews, S. Ramaswamy, F. Beaufays, S. Augenstein, et al., "Federated learning for mobile keyboard prediction," arXiv preprint arXiv:1811.03604, 2018.
[21] J. Konevcny, H. B. McMahan, F. X. Yu, P. Richtarik, A. T. Suresh, and D. Bacon, "Federated learning: Strategies for improving communication efficiency," arXiv preprint arXiv:1610.05492, 2016.
[22] NVIDIA, "什麼是聯合學習？" https://blogs.nvidia.com.tw/blog/what-is-federated-learning/, 2024.
[23] NVIDIA, "Introducing nvidia flare: An open federated learning framework." https://blogs.nvidia.com/blog/2021/05/04/introducing-nvidia-flare/, 2021.
[24] NVIDIA, "Nvidia flare developer guide." https://developer.nvidia.com/nvidia-flare, 2021.
[25] K. Bonawitz, H. Eichner, W. Grieskamp, D. Huba, A. Ingerman, V. Ivanov, et al., "Towards federated learning at scale: System design," in Proceedings of the 2nd SysML Conference, 2019.
[26] Y. Zhao, M. Li, L. Lai, N. Suda, D. Civin, and V. Chandra, "Federated learning with non-iid data," arXiv preprint arXiv:1806.00582, 2018.
[27] Q. Yang, Y. Liu, T. Chen, and Y. Tong, "Federated machine learning: Concept and applications," ACM Transactions on Intelligent Systems and Technology, vol. 10, no. 2, pp. 1-19, 2019.
[28] P. Ekman and W. V. Friesen, Facial Action Coding System (FACS): A Technique for the Measurement of Facial Action. 1978.
[29] J. F. Cohn, Z. Ambadar, and P. Ekman, Observer-Based Measurement of Facial Expression With the Facial Action Coding System, pp. 203-221. 2010.
[30] P. Ekman, "What the facial action coding system (facs) can tell us about facial expression," Emotion Researcher, 2020.
[31] T. Joachims, Making Large-Scale SVM Learning Practical, pp. 169-184. 1999.
[32] L. Breiman, "Random forests," Machine Learning, vol. 45, no. 1, pp. 5-32, 2001.
[33] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," Nature, vol. 521, no. 7553, pp. 436-444, 2015.
[34] M. Li, J. Deng, Z. Zhang, and G. Lu, "Deep learning for emotion recognition on small datasets using transfer learning," Proceedings of the 2019 International Conference on Multimodal Interaction, pp. 599-603, 2019.
[35] K. Zhang, Z. Zhang, Z. Li, and Y. Qiao, "Facial expression recognition based on deep evolutional spatial-temporal networks," IEEE Transactions on Image Processing, vol. 26, no. 9, pp. 4193-4203, 2017.
[36] A. Mollahosseini, D. Chan, and M. H. Mahoor, "Going deeper in facial expression recognition using deep neural networks," Proceedings of the IEEE Winter Conference on Applications of Computer Vision, pp. 1-10, 2016.
[37] C. A. Corneanu, M. Simón, J. F. Cohn, and S. E. Guerrero, "Survey on rgb, 3d, thermal, and multimodal approaches for facial expression recognition: History, trends, and affect-related applications," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 38, no. 8, pp. 1548-1568, 2016.
[38] B. Altinel, R. Hashimoto, Y. Yamaguchi, S. Sugino, and H. Akatsu, "Facial expression recognition using deep learning: A review," Symmetry, vol. 13, no. 6, pp. 1-25, 2021.
[39] B.-G. Kim and T.-J. Kim, "Facial expression recognition using convolutional neural networks: A comprehensive review," Applied Sciences, vol. 11, no. 2, p. 712, 2021.
[40] M. Agrawal et al., "Decentralized federated learning for eeg-based emotion recognition using 2d-cnn," IEEE Access, vol. 11, pp. 11234-11245, 2023.
[41] N. Simić et al., "Lightweight cnn for audio-visual emotion recognition on edge devices using federated learning," International Journal of Computer Vision, vol. 130, no. 5, pp. 1050-1064, 2022.
[42] A. R. Khan, "A review on facial emotion recognition techniques using traditional and deep learning approaches," IEEE Transactions on Affective Computing, vol. 13, no. 3, pp. 256-270, 2022.
[43] I. Goodfellow, D. Erhan, P. L. Carrier, A. Courville, M. Mirza, B. Hamner, et al., "Challenges in representation learning: A report on three machine learning contests," in Neural Information Processing Systems (NIPS), 2013.
[44] P. Lucey, J. F. Cohn, T. Kanade, J. Saragih, Z. Ambadar, and I. Matthews, "The extended cohn-kanade dataset (ck+): A complete dataset for action unit and emotion-specified expression," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, pp. 94-101, 2010.
[45] A. Mollahosseini, B. Hasani, and M. H. Mahoor, "Affectnet: A database for facial expression, valence, and arousal computing in the wild," IEEE Transactions on Affective Computing, vol. 10, no. 1, pp. 18-31, 2017.
[46] K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," arXiv preprint arXiv:1409.1556, 2014.
[47] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L. Chen, "Mobilenetv2: Inverted residuals and linear bottlenecks," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4510-4520, 2018.
[48] M. Abadi, A. Agarwal, P. Barham, E. Brevdo, Z. Chen, C. Citro, et al., "TensorFlow: Large-scale machine learning on heterogeneous systems." https://www.tensorflow.org/, 2015.
[49] A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, et al., "Pytorch: An imperative style, high-performance deep learning library," Advances in Neural Information Processing Systems, vol. 32, pp. 8024-8035, 2019.
[50] E. Union, "General data protection regulation (gdpr)." https://eur-lex.europa.eu/eli/reg/2016/679/oj, 2016.
[51] U. D. of Health & Human Services, "Health insurance portability and accountability act of 1996 (hipaa)." https://www.hhs.gov/hipaa/index.html, 1996.