Federated Learning (FL) is a promising distributed learning mechanism which still faces two major challenges, namely privacy breaches and system efficiency. In this work, we reconceptualize the FL system from the perspective of network information theory, and formulate an original FL communication framework, FedNC, which is inspired by Network Coding (NC). The main idea of FedNC is mixing the information of the local models by making random linear combinations of the original parameters, before uploading for further aggregation. Due to the benefits of the coding scheme, both theoretical and experimental analysis indicate that FedNC improves the performance of traditional FL in several important ways, including security, efficiency, and robustness. To the best of our knowledge, this is the first framework where NC is introduced in FL. As FL continues to evolve within practical network frameworks, more variants can be further designed based on FedNC.

Federated Learning (FL) is a promising distributed learning mechanism that revolutionizes our interaction with data in the IoT ecosystem. Due to the rapidly growing scale of smart devices and the limited transmission resources of networks, a simple, consistent and scalable FL framework aiming to address the communication bottleneck is urgently needed. In this work, we propose an efficient approach with Selective Aggregation of Models (SAM) to mitigate the communication overload in FL systems. The introduction of SAM enables each local client to upload its model with a certain probability, resulting in a significant reduction in costly communication expenses. We design the algorithm for SAM, analyze the convergence bound on non-convex objectives for heterogeneous data, which illustrates the impact of the selection probability as well as the set size of participating clients on the system performance, and assess the conservation for the network resource utilization by modeling queuing systems. We conduct various experiments to evaluate the performance of SAM, whose outcomes suggest that significant alleviation of the communication bottleneck can be accomplished with marginal cost of performance loss. It will also be shown that SAM is a communication-efficient method that can be freely applied to other frameworks.

As a promising learning paradigm integrating computation and communication, federated learning (FL) proceeds the local training and the periodic sharing from distributed clients. Due to the non-i.i.d. data distribution on clients, FL model suffers from the gradient diversity, poor performance, bad convergence, etc. In this work, we aim to tackle this key issue by adopting importance sampling (IS) for local training. We propose importance sampling federated learning (ISFL), an explicit framework with theoretical guarantees. Firstly, we derive the convergence theorem of ISFL to involve the effects of local importance sampling. Then, we formulate the problem of selecting optimal IS weights and obtain the theoretical solutions. We also employ a water-filling method to calculate the IS weights and develop the ISFL algorithms. The experimental results on CIFAR-10 fit the proposed theorems well and verify that ISFL reaps better performance, convergence, sampling efficiency, as well as explainability on non-i.i.d. data. To the best of our knowledge, ISFL is the first non-i.i.d. FL solution from the local sampling aspect which exhibits theoretical compatibility with neural network models. Furthermore, as a local sampling approach, ISFL can be easily migrated into other emerging FL frameworks.

With the development of large-scale intelligent services, semantic communication has attracted significant interest from both academia and industry, which is expected to transmit valuable data traffic at sufficiently high speed with extremely low end-to-end latency. However, the generation and measurement of semantic messages is still an open problem. On the other hand, expansion which combines simple things into complex systems and even generates intelligence, is consistent with the evolution of human civilization and language systems. Motivated by this key idea, we apply it to semantic communication systems, measuring semantics carried by symbol sequences, and similarly investigate the semantic information system as Shannon did for digital communication systems. This work was the first to propose the concept of semantic expansion and knowledge collision, which may provide a new paradigm for semantic communications. We believe that expansion and collision will be the cornerstone of semantic information theory.

Federated learning (FL) has prevailed as an efficient and privacy-preserved scheme for distributed learning. In this work, we mainly focus on the optimization of computation and communication in FL from a view of pruning. By adopting layer-wise pruning in local training and federated updating, we formulate an explicit FL pruning framework, FedLP (Federated Layer-wise Pruning), which is model-agnostic and universal for different types of deep learning models. Two specific schemes of FedLP are designed for scenarios with homogeneous local models and heterogeneous ones. Both theoretical and experimental evaluations are developed to verify that FedLP relieves the system bottlenecks of communication and computation with marginal performance decay. To the best of our knowledge, FedLP is the first framework that formally introduces the layer-wise pruning into FL. Within the scope of federated learning, more variants and combinations can be further designed based on FedLP.

With the emergence of communication systems and deep learning techniques, the native intelligence has been envisioned as a primary power of future networks. In this work, we investigate the schemes of distributed communication-computation integrated networks and propose a split learning based solution for multi-gNB intelligence, abbreviated as MgCSL. By carrying out a data-model split mechanism, MgC-SL mitigates the computation requirements of each node and enables more gNBs to participate the collaborative learning tasks. The simulation results verify that such distributed scheme significantly saves the communication and computation costs without the degradation of the task performance. A joint indicator is also formulated for performance analysis. Combining the proposed schemes and the corresponding indicator, some insights and guides for the system designs can be obtained to improve the efficiency of the next-generation network intelligence.

Inspired by diversity technology, we rethink the model enhancement from the view of wireless communication and propose a space-time framework for ensemble learning, called diversity learning. Such framework provides a new perspective that links the multi-model learning with the multi-channel commu-nication. In this paper, 2×1 diversity learning is mainly studied whose efficiency is guaranteed theoretically. We also evaluate the proposed scheme on two popular image classification tasks, MNIST and CIFAR-10. The results elucidate that the diversity learning reaps superiority on model enhancement, convergence, complexity and robustness compared to single models as well as weighting ensemble approach. Furthermore, the diversity schemes can be deployed in several emerging distributed learning systems, especially the mobile scenarios such as edge computing and cooperative learning where the resources for computation and communication are restricted.

As an emerging technique, mobile-edge computing (MEC) introduces a new scheme for various distributed communication-computing systems, such as industrial Internet of Things (IoT), vehicular communication, smart city, etc. In this work, we mainly focus on the timeliness of the MEC systems where the freshness of the data and computation tasks is significant. First, we formulate a kind of age-sensitive MEC models and define the average Age-of-Information (AoI) minimization problems of interests. Then, a novel mixed-policy-based multimodal deep reinforcement learning (RL) framework, called heterogeneous multiagent actor–critic (H-MAAC), is proposed as a paradigm for joint collaboration in the investigated MEC systems, where edge devices and center controller learn the interactive strategies through their own observations. To improve the system performance, we develop the corresponding online algorithm by introducing the edge federated learning mode into the multiagent cooperation whose advantages on learning convergence can be guaranteed theoretically. To the best of our knowledge, it is the first joint MEC collaboration algorithm that combines the edge federated mode with the multiagent actor–critic RL. Furthermore, we evaluate the proposed approach and compare it with popular RL-based methods. As a result, the proposed algorithm not only outperforms the baselines on average system age, but also promotes the stability of training process. Besides, the simulation outcomes provide several insights for collaboration designs over MEC systems.

With the rapid growth of the data volume and the fast increasing of the computational model complexity in the scenario of cloud computing, it becomes an important topic that how to handle users' requests by scheduling computational jobs and assigning the resources in data center.In order to have a better perception of the computing jobs and their requests of resources, we analyze its characteristics and focus on the prediction and classification of the computing jobs with some machine learning approaches. Specifically, we apply LSTM neural network to predict the arrival of the jobs and the aggregated requests for computing resources. Then we evaluate it on Google Cluster dataset and it shows that the accuracy has been improved compared to the current existing methods. Additionally, to have a better understanding of the computing jobs, we use an unsupervised hierarchical clustering algorithm, BIRCH, to make classification and get some interpretability of our results in the computing centers.