Fuelled by the availability of more data and computing power, recent breakthroughs in cloud-based machine learning (ML) have transformed every aspect of our lives from face recognition and medical diagnosis to natural language processing. However, classical ML exerts severe demands in terms of energy, memory, and computing resources, limiting their adoption for resource constrained edge devices. The new breed of intelligent devices requires a novel paradigm change calling for distributed, low-latency, and reliable ML at the wireless network edge. In this case, building comprehensive artificial intelligent (AI) platform capabilities gives rise to new challenges, such as: (i) how to orchestrate and manage heterogeneous and distributed resources? (ii) how to define a universal and efficient mechanism to provide diversified AI services, which include perception, data mining, prediction, and reasoning? (iii) how to leverage the 6G system capability and highly reliable AI services? In addition, semantic and goal-oriented communications are seen as a promising technology going beyond transferring and reconstructing information bits (Shannon’s level-A). By leveraging the agents’ capabilities of abstraction and reasoning, semantic communication has the potential to make wireless networks significantly more efficient, reliable, resilient, and sustainable. The advancements of machine learning (ML) make such objectives promising, where significant work arises recently in both theories and applications. To build the pathways towards semantic-native communications, some fundamental problems and open challenges should be studied carefully. This includes but not limited to, defining mechanisms of semantic abstraction and reasoning, bringing cognition to the communication system, and emerging robust, generalized, resilient communication protocols to wireless scenarios.
This workshop aims to bring together practitioners and researchers from both academia and industry for discussions and technical presentations on fundamental and practically relevant questions related to the many challenges arising from Native AI and the semantic communications paradigm. Topics of interest include, but are not limited to:
- Concept, architectures, theories and applications of AI for Network and Network for AI
- Large Language Model (LLM) for native AI network and semantic communications;
- Multi-agent reinforcement learning for native AI network and semantic communications;
- AI/ML empowered communication system and protocol design, such as physical layer, signal processing, source and channel coding, radio resource management;
- Distributed or federated learning framework empowered by communication networks;
- AI/ML empowered wireless sensing, imaging, localization;
- AI/ML empowered multi-modality sensors (camera, LiDAR, radar, GPS) to assist wireless communications, such as beamforming, handover, power allocation;
- AI/ML empowered semantic, emergent, goal oriented communications;
- Theory of semantic information, representation, and abstraction;
- Knowledge and reasoning driven semantic communication networks, including system 2 ML concepts such as causality, belief transport, propositional logic;
- Applications of native AI networks and semantic communications: conversation, query, natural language understanding, visual recognition, generation, etc.
- Vertical use cases of native AI networks and semantic communications: autonomous driving, extended reality, smart manufacturing, etc.
- Experiments, prototypes, demonstrations of native AI networks and semantic communications.
EDAS Submission Link: https://www.edas.info/newPaper.php?c=30964&track=118019
Merouane Debbah, TII, UAE
Ebtesam Almazrouei, TII, UAE
Daniel Benevides da Costa, TII, UAE
Qiyang Zhao, TII, UAE
Paper Submission Deadline:
19 May 2023 Extended date: 31 May 2023
Acceptance Notification: 16 June 2023
Camera-Ready Due: 07 July 2023
Workshop Date: 05 September 2023