Tensor-Based Efficient Federated Reinforcement Learning for Cyber-Physical-Social Intelligence
Reinforcement Learning (RL) serves as a fundamental learning paradigm in the field of artificial intelligence, enabling decision-making policies through interactions with environments. However, traditional RL methods encounter challenges when dealing with large-scale or continuous state spaces due t...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Tsinghua University Press
2025-06-01
|
| Series: | Big Data Mining and Analytics |
| Subjects: | |
| Online Access: | https://www.sciopen.com/article/10.26599/BDMA.2025.9020036 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849304042008739840 |
|---|---|
| author | Xin Nie Laurence T. Yang Fulan Fan Zecan Yang |
| author_facet | Xin Nie Laurence T. Yang Fulan Fan Zecan Yang |
| author_sort | Xin Nie |
| collection | DOAJ |
| description | Reinforcement Learning (RL) serves as a fundamental learning paradigm in the field of artificial intelligence, enabling decision-making policies through interactions with environments. However, traditional RL methods encounter challenges when dealing with large-scale or continuous state spaces due to the curse of dimensionality. Although Deep Reinforcement Learning (DRL) can handle complex environments, its lack of transparency and interpretability hinders its applicability due to the black box nature. Moreover, centralized data collection and processing methods pose privacy security risks. Federated learning offers a distributed approach that ensures privacy preservation while co-training models. However, existing federated reinforcement learning approaches have not adequately addressed communication and computation overhead issues. To address these challenges, this study proposes a tensor train decomposition-based federated reinforcement learning method that enhances efficiency and provides interpretability. By leveraging tensor to model state-action values and employing tensor decomposition techniques for dimensionality reduction, this method effectively reduces model parameters and communication overhead while maintaining strong interpretability, accelerates algorithm convergence speed. Experimental results validate the advantages of our proposed algorithm in terms of efficiency and reliability. |
| format | Article |
| id | doaj-art-a9f012c9eded454d896f0b4e0a859c72 |
| institution | Kabale University |
| issn | 2096-0654 2097-406X |
| language | English |
| publishDate | 2025-06-01 |
| publisher | Tsinghua University Press |
| record_format | Article |
| series | Big Data Mining and Analytics |
| spelling | doaj-art-a9f012c9eded454d896f0b4e0a859c722025-08-20T03:55:49ZengTsinghua University PressBig Data Mining and Analytics2096-06542097-406X2025-06-018486787910.26599/BDMA.2025.9020036Tensor-Based Efficient Federated Reinforcement Learning for Cyber-Physical-Social IntelligenceXin Nie0Laurence T. Yang1Fulan Fan2Zecan Yang3School of Computer and Artiffcial Intelligence, Zhengzhou University, Zhengzhou 450001, ChinaSchool of Computer and Artificial Intelligence, Zhengzhou University, Zhengzhou 450001, China, and also with Department of Computer Science, St. Francis Xavier University, Antigonish, B2G 2W5, CanadaSchool of Education, South-Central Minzu University, Wuhan 430074, ChinaSchool of Computer and Artiffcial Intelligence, Zhengzhou University, Zhengzhou 450001, ChinaReinforcement Learning (RL) serves as a fundamental learning paradigm in the field of artificial intelligence, enabling decision-making policies through interactions with environments. However, traditional RL methods encounter challenges when dealing with large-scale or continuous state spaces due to the curse of dimensionality. Although Deep Reinforcement Learning (DRL) can handle complex environments, its lack of transparency and interpretability hinders its applicability due to the black box nature. Moreover, centralized data collection and processing methods pose privacy security risks. Federated learning offers a distributed approach that ensures privacy preservation while co-training models. However, existing federated reinforcement learning approaches have not adequately addressed communication and computation overhead issues. To address these challenges, this study proposes a tensor train decomposition-based federated reinforcement learning method that enhances efficiency and provides interpretability. By leveraging tensor to model state-action values and employing tensor decomposition techniques for dimensionality reduction, this method effectively reduces model parameters and communication overhead while maintaining strong interpretability, accelerates algorithm convergence speed. Experimental results validate the advantages of our proposed algorithm in terms of efficiency and reliability.https://www.sciopen.com/article/10.26599/BDMA.2025.9020036cyber-physical-social intelligence (cpsi)federated reinforcement learning (frl)tensor train decomposition |
| spellingShingle | Xin Nie Laurence T. Yang Fulan Fan Zecan Yang Tensor-Based Efficient Federated Reinforcement Learning for Cyber-Physical-Social Intelligence Big Data Mining and Analytics cyber-physical-social intelligence (cpsi) federated reinforcement learning (frl) tensor train decomposition |
| title | Tensor-Based Efficient Federated Reinforcement Learning for Cyber-Physical-Social Intelligence |
| title_full | Tensor-Based Efficient Federated Reinforcement Learning for Cyber-Physical-Social Intelligence |
| title_fullStr | Tensor-Based Efficient Federated Reinforcement Learning for Cyber-Physical-Social Intelligence |
| title_full_unstemmed | Tensor-Based Efficient Federated Reinforcement Learning for Cyber-Physical-Social Intelligence |
| title_short | Tensor-Based Efficient Federated Reinforcement Learning for Cyber-Physical-Social Intelligence |
| title_sort | tensor based efficient federated reinforcement learning for cyber physical social intelligence |
| topic | cyber-physical-social intelligence (cpsi) federated reinforcement learning (frl) tensor train decomposition |
| url | https://www.sciopen.com/article/10.26599/BDMA.2025.9020036 |
| work_keys_str_mv | AT xinnie tensorbasedefficientfederatedreinforcementlearningforcyberphysicalsocialintelligence AT laurencetyang tensorbasedefficientfederatedreinforcementlearningforcyberphysicalsocialintelligence AT fulanfan tensorbasedefficientfederatedreinforcementlearningforcyberphysicalsocialintelligence AT zecanyang tensorbasedefficientfederatedreinforcementlearningforcyberphysicalsocialintelligence |