Federated Learning With Heterogeneity-Aware Probabilistic Synchronous Parallel on Edge

Article Computer Science, Theory & Methods

Self-Balancing Federated Learning With Global Imbalanced Data in Mobile Systems

Moming Duan et al.

Summary: Federated Learning (FL) is a distributed deep learning method where multiple devices contribute to a neural network training while keeping their data private. Data imbalance in mobile systems can lead to accuracy degradation in FL applications, but the Astraea framework offers improvements through data augmentation and rescheduling. Compared to FedAvg, Astraea demonstrates higher accuracy and reduced communication traffic.

IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS (2021)

添加到收藏夹

Article Computer Science, Information Systems

NOMA-Enabled Cooperative Computation Offloading for Blockchain-Empowered Internet of Things: A Learning Approach

Zhenni Li et al.

Summary: This article introduces a multiagent deep reinforcement learning framework to address the cooperative computation offloading problem, by using scatter networks and league learning to improve stability and performance.

IEEE INTERNET OF THINGS JOURNAL (2021)

添加到收藏夹

Article Computer Science, Theory & Methods

Distributed and Collective Deep Reinforcement Learning for Computation Offloading: A Practical Perspective

Xiaoyu Qiu et al.

Summary: DC-DRL is a distributed and collective DRL algorithm proposed to address the gap in deploying real DRL applications in MEC. By assimilating knowledge from multiple MEC environments and using adaptive n-step learning, DC-DRL effectively increases data diversity, reduces exploration costs, and maximizes the utilization of multiple environments to prevent premature convergence. Evaluation results show a significant reduction in exploration costs and final system costs compared to baselines.

IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Communication-Efficient and Distributed Learning Over Wireless Networks: Principles and Applications

Jihong Park et al.

Summary: The article discusses the application of machine learning in 5G communication systems, emphasizing the importance of injecting intelligence at the network edge to optimize communication systems, and how improving communication efficiency through distributed learning is key.

PROCEEDINGS OF THE IEEE (2021)

添加到收藏夹

Article Computer Science, Information Systems