Nikos A. Mitsiou

Nikos G. Evgenidis, Nikos A. Mitsiou, Vasiliki I. Koutsioumpa et al.

This paper focuses on the latest research and innovations in fundamental next-generation multiple access (NGMA) techniques and the coexistence with other key technologies for the sixth generation (6G) of wireless networks. In more detail, we first examine multi-access edge computing (MEC), which is critical to meeting the growing demand for data processing and computational capacity at the edge of the network, as well as network slicing. We then explore over-the-air (OTA) computing, which is considered to be an approach that provides fast and efficient computation of various functions. We also explore semantic communications, identified as an effective way to improve communication systems by focusing on the exchange of meaningful information, thus minimizing unnecessary data and increasing efficiency. The interrelationship between machine learning (ML) and multiple access technologies is also reviewed, with an emphasis on federated learning, federated distillation, split learning, reinforcement learning, and the development of ML-based multiple access protocols. Finally, the concept of digital twinning and its role in network management is discussed, highlighting how virtual replication of physical networks can lead to improvements in network efficiency and reliability.

Nikos G. Evgenidis, Nikos A. Mitsiou, Sotiris A. Tegos et al.

In this paper, we propose a novel approach to minimize the inference delay in semantic segmentation using split learning (SL), tailored to the needs of real-time computer vision (CV) applications for resource-constrained devices. Semantic segmentation is essential for applications such as autonomous vehicles and smart city infrastructure, but faces significant latency challenges due to high computational and communication loads. Traditional centralized processing methods are inefficient for such scenarios, often resulting in unacceptable inference delays. SL offers a promising alternative by partitioning deep neural networks (DNNs) between edge devices and a central server, enabling localized data processing and reducing the amount of data required for transmission. Our contribution includes the joint optimization of bandwidth allocation, cut layer selection of the edge devices' DNN, and the central server's processing resource allocation. We investigate both parallel and serial data processing scenarios and propose low-complexity heuristic solutions that maintain near-optimal performance while reducing computational requirements. Numerical results show that our approach effectively reduces inference delay, demonstrating the potential of SL for improving real-time CV applications in dynamic, resource-constrained environments.

Nikos A. Mitsiou, Pavlos S. Bouzinis, Panagiotis G. Sarigiannidis et al.

The optimal solution to an optimization problem depends on the problem's objective function, constraints, and size. While deep neural networks (DNNs) have proven effective in solving optimization problems, changes in the problem's size, objectives, or constraints often require adjustments to the DNN architecture to maintain effectiveness, or even retraining a new DNN from scratch. Given the dynamic nature of wireless networks, which involve multiple and diverse objectives that can have conflicting requirements and constraints, we propose a multi-task learning (MTL) framework to enable a single DNN to jointly solve a range of diverse optimization problems. In this framework, optimization problems with varying dimensionality values, objectives, and constraints are treated as distinct tasks. To jointly address these tasks, we propose a conditional computation-based MTL approach with routing. The multi-task DNN consists of two components, the base DNN (bDNN), which is the single DNN used to extract the solutions for all considered optimization problems, and the routing DNN (rDNN), which manages which nodes and layers of the bDNN to be used during the forward propagation of each task. The output of the rDNN is a binary vector which is multiplied with all bDNN's weights during the forward propagation, creating a unique computational path through the bDNN for each task. This setup allows the tasks to either share parameters or use independent ones, with the decision controlled by the rDNN. The proposed framework supports both supervised and unsupervised learning scenarios. Numerical results demonstrate the efficiency of the proposed MTL approach in solving diverse optimization problems. In contrast, benchmark DNNs lacking the rDNN mechanism were unable to achieve similar levels of performance, highlighting the effectiveness of the proposed architecture.