FedDyMem: Efficient Federated Learning with Dynamic Memory and Memory-Reduce for Unsupervised Image Anomaly Detection

TL;DR

FedDyMem introduces a federated learning approach for unsupervised image anomaly detection that uses dynamic memory banks and memory reduction techniques to enhance privacy, efficiency, and performance across diverse industrial and medical datasets.

Contribution

The paper proposes FedDyMem, a novel federated learning method utilizing dynamic memory banks and memory reduction to improve unsupervised image anomaly detection while preserving data privacy.

Findings

Effective across six industrial and medical datasets.

Significantly reduces memory size for communication efficiency.

Enhances anomaly detection accuracy in federated settings.

Abstract

Unsupervised image anomaly detection (UAD) has become a critical process in industrial and medical applications, but it faces growing challenges due to increasing concerns over data privacy. The limited class diversity inherent to one-class classification tasks, combined with distribution biases caused by variations in products across and within clients, poses significant challenges for preserving data privacy with federated UAD. Thus, this article proposes an efficient federated learning method with dynamic memory and memory-reduce for unsupervised image anomaly detection, called FedDyMem. Considering all client data belongs to a single class (i.e., normal sample) in UAD and the distribution of intra-class features demonstrates significant skewness, FedDyMem facilitates knowledge sharing between the client and server through the client's dynamic memory bank instead of model parameters. In the local clients, a memory generator and a metric loss are employed to improve the consistency of the feature distribution for normal samples, leveraging the local model to update the memory bank dynamically. For efficient communication and data privacy, a memory-reduce method based on weighted averages is proposed to significantly decrease the scale of memory banks. This reduced representation inherently, thereby mitigating the risk of data reconstruction. On the server, global memory is constructed and distributed to individual clients through k-means aggregation. Experiments conducted on six industrial and medical datasets, comprising a mixture of six products or health screening types derived from eleven public datasets, demonstrate the effectiveness of FedDyMem.