Task-Agnostic Semantic Communications Relying on Information Bottleneck and Federated Meta-Learning
- 24 November 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Wireless Communications
- Vol. 25 (15361276) , 7600-7616
Abstract
As a paradigm shift towards pervasive intelligence, semantic communication (SemCom) has shown great potentials to improve communication efficiency and provide user-centric services by delivering task-oriented semantic meanings. However, the exponential growth in connected devices, data volumes, and communication demands presents significant challenges for practical SemCom design, particularly in resource-constrained wireless networks. In this work, we propose a task-agnostic semantic communication (TASC) framework capable of supporting multimodal data across diverse tasks. To investigate the interplay between communication and intelligent tasks from an information-theoretic perspective, we introduce a distributed multimodal information bottleneck (DMIB) principle, which enables the extraction of minimal sufficient unimodal and multimodal representations by eliminating redundant information while preserving task-relevant semantics. To further reduce the communication overhead, we develop an adaptive semantic feature transmission method under dynamic channel conditions. Then, TASC is trained based on federated meta-learning (FML) to learn a well-initialized model for rapid adaptation and generalization. To gain deep insights, we conduct theoretical analysis and devise resource management to accelerate convergence while minimizing the training latency and energy cost. Moreover, we develop a joint user selection and resource allocation algorithm to address the non-convex problem with theoretical guarantees. Extensive simulation results validate the effectiveness and superiority of the proposed TASC compared to baselines.Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (62293485)
- National Natural Science Foundation of China (62501655)
- Natural Science Foundation of Jiangsu Province (BK20250291)
- Jiangsu Funding Program for Excellent Postdoctoral Talent
- Seatrium New Energy Laboratory, Singapore Ministry of Education (MOE) Tier 1 (RT5/23)
- Seatrium New Energy Laboratory, Singapore Ministry of Education (MOE) Tier 1 (RG24/24)
- Nanyang Technological University (NTU) Centre for Computational Technologies in Finance
- Research Innovation and Enterprise (RIE) 2025 Industry Alignment Fund-Industry Collaboration Projects (IAF-ICP) administered by the Agency for Science, Technology and Research (I2301E0026)
- China Scholarship Council (I2301E0026)
This publication has 31 references indexed in Scilit:
- When AI meets sustainable 6GScience China Information Sciences, 2024
- Federated Knowledge Distillation Enabled Image Semantic Communication2024
- Intellicise Wireless Networks From Semantic Communications: A Survey, Research Issues, and ChallengesIEEE Communications Surveys & Tutorials, 2024
- Transitioning From Federated Learning to Quantum Federated Learning in Internet of Things: A Comprehensive SurveyIEEE Communications Surveys & Tutorials, 2024
- Federated Learning and Meta Learning: Approaches, Applications, and DirectionsIEEE Communications Surveys & Tutorials, 2023
- Task-Oriented Multi-User Semantic CommunicationsIEEE Journal on Selected Areas in Communications, 2022
- Distributed Image Transmission Using Deep Joint Source-Channel Coding2022
- Toward Wisdom-Evolutionary and Primitive-Concise 6G: A New Paradigm of Semantic Communication NetworksEngineering, 2022
- Federated Learning based Audio Semantic Communication over Wireless Networks2021
- UniT: Multimodal Multitask Learning with a Unified Transformer2021