[论文解读] BrainTorrent: A Peer-to-Peer Environment for Decentralized Federated Learning

Abhijit Guha Roy, Shayan Siddiqui+3 more|arXiv (Cornell University)|May 16, 2019

Privacy-Preserving Technologies in Data参考文献 8被引用 207

一句话总结

BrainTorrent 提供了一个完全去中心化的点对点联邦学习框架用于医学影像，在没有中央服务器的情况下实现了有竞争力的全脑分割，并在不同数据分布下超过基于服务器的联邦学习。

ABSTRACT

Access to sufficient annotated data is a common challenge in training deep neural networks on medical images. As annotating data is expensive and time-consuming, it is difficult for an individual medical center to reach large enough sample sizes to build their own, personalized models. As an alternative, data from all centers could be pooled to train a centralized model that everyone can use. However, such a strategy is often infeasible due to the privacy-sensitive nature of medical data. Recently, federated learning (FL) has been introduced to collaboratively learn a shared prediction model across centers without the need for sharing data. In FL, clients are locally training models on site-specific datasets for a few epochs and then sharing their model weights with a central server, which orchestrates the overall training process. Importantly, the sharing of models does not compromise patient privacy. A disadvantage of FL is the dependence on a central server, which requires all clients to agree on one trusted central body, and whose failure would disrupt the training process of all clients. In this paper, we introduce BrainTorrent, a new FL framework without a central server, particularly targeted towards medical applications. BrainTorrent presents a highly dynamic peer-to-peer environment, where all centers directly interact with each other without depending on a central body. We demonstrate the overall effectiveness of FL for the challenging task of whole brain segmentation and observe that the proposed server-less BrainTorrent approach does not only outperform the traditional server-based one but reaches a similar performance to a model trained on pooled data.

研究动机与目标

• 解决医学影像联邦学习中的数据隐私和对中央服务器的依赖问题。
• 开发适用于具有多样数据分布的医疗中心的无服务器点对点FL框架。
• 证明 BrainTorrent 在保持数据隐私的同时，分割性能可与合并数据相媲美。

提出的方法

• 将 BrainTorrent 引入为一个无服务器的点对点FL环境。
• 每个客户端维护一个版本向量来跟踪模型更新。
• 一个随机客户端发起轮次；更新的同伴发送它们的权重和数据量以形成合并模型。
• 合并模型在本地数据上进行微调，使得在没有中央服务器的情况下快速收敛。
• 将 BrainTorrent 与传统的基于服务器的FL（FLS）以及合并数据训练在全脑MRI分割上的表现进行比较。

实验结果

研究问题

• RQ1 一个去中心化、无服务器的FL框架是否能够在多中心实现与合并数据相当的分割性能？
• RQ2 BrainTorrent 是否对中心间的非均匀和按年龄分层的数据分布具有鲁棒性？
• RQ3 随着客户端数量增加和每个客户端数据量减少，BrainTorrent 的扩展性如何？
• RQ4 对于数据集非常小的中心，BrainTorrent 是否比基于服务器的FL更能保留性能？

主要发现

• 在所有测试配置中，BrainTorrent 在客户端的平均 Dice 分数和聚合模型上均优于基于服务器的FL。
• 随着客户端数量增加（每个客户端数据减少），BrainTorrent 保持优于 FLS 的性能并接近合并数据的性能。
• 在非均匀数据分布场景下，BrainTorrent 的聚合和按客户端的性能接近或接近合并模型，而 FLS 会下降。
• 按客户端的分析显示 BrainTorrent 能为大多数中心带来更鲁棒的个性化模型，在 10 客户端设置下平均 Dice 提升约 2 个百分点，超过 FLS。
• BrainTorrent 即使不共享原始数据，也达到与基于合并数据训练的模型相当的分割精度。