arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2204.04424cs.LGcs.AIcs.CVcs.DC
收藏

用于快速和通信高效的联邦学习的自适应差分滤波器

Adaptive Differential Filters for Fast and Communication-Efficient Federated Learning

Fraunhofer Heinrich Hertz Institute (HHI)(弗劳恩霍夫海因里希·赫兹研究所(HHI))

Daniel Becking, Heiner Kirchhoffer, Gerhard Tech, Paul Haase, Karsten Müller, Heiko Schwarz, Wojciech Samek

摘要:

联邦学习(FL)场景通过频繁地在客户端和服务器之间传输神经网络更新而固有地产生大量通信开销。为了最小化通信成本,引入稀疏性结合差分更新是一种常用的技术。然而,稀疏模型更新可能会减慢收敛速度或无意中跳过某些更新方面,例如,如果错误积累没有得到适当解决,则会导致学习到的特征。在这项工作中,我们提出了一种在卷积滤波器粒度下操作的新缩放方法,该方法1)补偿FL过程中的高度稀疏更新,2)通过增强滤波器空间中的一些特征同时减少其他特征来使局部模型适应新的数据域,以及3)激励更新中的额外稀疏性,从而实现更高的压缩比,即,节省整个数据传输。与未缩放的更新和以前的工作相比,在单向,双向和部分更新FL设置中的不同计算机视觉任务(Pascal VOC,CIFAR10,胸部X射线)和神经网络(ResNets,MobileNets,VGG)上的实验结果表明,所提出的方法提高了中央服务器模型的性能,同时收敛速度更快,并将传输的数据总量减少了377倍。

英文摘要:

Federated learning (FL) scenarios inherently generate a large communication overhead by frequently transmitting neural network updates between clients and server. To minimize the communication cost, introducing sparsity in conjunction with differential updates is a commonly used technique. However, sparse model updates can slow down convergence speed or unintentionally skip certain update aspects, e.g., learned features, if error accumulation is not properly addressed. In this work, we propose a new scaling method operating at the granularity of convolutional filters which 1) compensates for highly sparse updates in FL processes, 2) adapts the local models to new data domains by enhancing some features in the filter space while diminishing others and 3) motivates extra sparsity in updates and thus achieves higher compression ratios, i.e., savings in the overall data transfer. Compared to unscaled updates and previous work, experimental results on different computer vision tasks (Pascal VOC, CIFAR10, Chest X-Ray) and neural networks (ResNets, MobileNets, VGGs) in uni-, bidirectional and partial update FL settings show that the proposed method improves the performance of the central server model while converging faster and reducing the total amount of transmitted data by up to 377 times.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑