联邦生成模型中自适应加权客户端，提升医学影像生成质量
FedCAR: Cross-client Adaptive Re-weighting for Generative Models in Federated Learning
- 根据客户端生成假图像的分布差异动态调整贡献权重
- 在三个胸部X光数据集上优于集中式学习和传统联邦学习
- 适合医疗领域跨机构隐私保护下的生成模型训练
在多机构数据上训练生成模型可借助多样化的数据分布获得更丰富的理解。然而，由于医院出于隐私考虑不愿共享数据，医学图像的模型训练面临挑战。联邦学习（FL）通过聚合多个客户端的模型参数而非共享原始数据，成为一种保护隐私的分布式训练方案。尽管已有研究探索将FL应用于生成模型，但针对生成模型的高效聚合算法仍不充分。本文提出一种新算法，通过自适应重加权各客户端贡献，实现高质量共享参数的训练。每轮迭代中，服务器端测量客户端生成假图像之间的分布距离，而非逐个比较弗雷切特起始距离（FID），显著提升效率。在三个公开胸部X光数据集上的实验表明，该方法在医学图像生成任务中表现优异，超越集中式学习与传统联邦学习算法。代码已开源：https://github.com/danny0628/FedCAR。
原文摘要 · Abstract (English)
Generative models trained on multi-institutional datasets can provide an enriched understanding through diverse data distributions. However, training the models on medical images is often challenging due to hospitals' reluctance to share data for privacy reasons. Federated learning(FL) has emerged as a privacy-preserving solution for training distributed datasets across data centers by aggregating model weights from multiple clients instead of sharing raw data. Previous research has explored the adaptation of FL to generative models, yet effective aggregation algorithms specifically tailored for generative models remain unexplored. We hereby propose a novel algorithm aimed at improving the performance of generative models within FL. Our approach adaptively re-weights the contribution of each client, resulting in well-trained shared parameters. In each round, the server side measures the distribution distance between fake images generated by clients instead of directly comparing the Fréchet Inception Distance per client, thereby enhancing efficiency of the learning. Experimental results on three public chest X-ray datasets show superior performance in medical image generation, outperforming both centralized learning and conventional FL algorithms. Our code is available at https://github.com/danny0628/FedCAR.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。