Abstract:Large amounts of data is necessary for deep learning models to semantically segment images. A major issue in the field of medical imaging is accumulating adequate data and then applying specialized skills to label those medical imaging data. Collaboration across institutions might be able to alleviate this problem, but sharing medical data to a centralized place is complicated due to legal, privacy, technical, and data ownership constraints, particularly among international institutions. By guaranteeing user privacy and preventing unauthorized access to raw data, Federated Learning plays a significant role especially in decentralized deep learning applications. Each client is given a unique learning process assignment. Clients first train a machine learning model locally using data from their area. Then, clients upload training data (local updates of model weights and biases) to a server. After that, the server compiles client‐provided updates to build a global learning model. Due to the numerous parameters (weights and biases) employed by deep learning models, the constant transmission between clients and the server raises communication costs and is inefficient from the standpoint of resource use. When there are more contributing clients and communication rounds, the cost of communication becomes a bigger concern. In this paper, a novel federated learning with weight sharing optimization compression architecture FedWSOcomp is proposed for cross institutional collaboration. In FedWSOcomp, the weights from deep learning models between clients and servers help in considerably reducing the amount of updates. Top‐z sparsification, quantization with clustering, and compression with three separate encoding techniques are all implemented by FedWSOcomp. Modern compression techniques are outperformed by FedWSOcomp, which achieves compression rates of up to 1085× while saving up to 99% of bandwidth and 99% of energy for clients during communication.

Federated Neural Compression Under Heterogeneous Data

FedPSE: Personalized Sparsification with Element-wise Aggregation for Federated Learning

Efficient Model Compression for Hierarchical Federated Learning

Secure Federated Learning with Model Compression.

Wyner-Ziv Gradient Compression for Federated Learning

FedConv: A Learning-on-Model Paradigm for Heterogeneous Federated Clients

Exploiting Shared Representations for Personalized Federated Learning

Adaptive Compression in Federated Learning via Side Information

Bandwidth-Aware and Overlap-Weighted Compression for Communication-Efficient Federated Learning

Compression with Exact Error Distribution for Federated Learning

Learning to Generalize in Heterogeneous Federated Networks

FedSZ: Leveraging Error-Bounded Lossy Compression for Federated Learning Communications

Federated Learning with Data-Agnostic Distribution Fusion

Deep learning based brain tumour architecture for weight sharing optimization in federated learning

Federated Learning Model Aggregation in Heterogenous Aerial and Space Networks

Scalable Hybrid Learning Techniques for Scientific Data Compression

Stochastic gradient compression for federated learning over wireless network

Deep Hierarchy Quantization Compression algorithm based on Dynamic Sampling

Privacy-Aware Compression for Federated Data Analysis

Accelerated Methods with Compression for Horizontal and Vertical Federated Learning

Factor-Assisted Federated Learning for Personalized Optimization with Heterogeneous Data