arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1909.04716cs.LGcs.DCcs.NAmath.NAmath.OCstat.ML
收藏

压缩迭代的梯度下降

Gradient Descent with Compressed Iterates

Cairo University(开罗大学);KAUST(沙特国王 Abdullah 科技大学)

Ahmed Khaled, Peter Richtárik

摘要:

提出并分析了一种新的随机一阶方法:压缩迭代梯度下降(GDCI)。GDCI在每次迭代中首先使用有损随机化压缩技术压缩当前帧,随后采取梯度步骤。这种方法是对当前联邦学习实践中一个关键成分的提炼,在联邦学习中,模型需要在发送回服务器进行聚合之前由移动终端进行压缩。我们的分析为缩小联邦学习的理论和实践之间的差距迈出了一步,并为许多扩展提供了可能性。

英文摘要:

We propose and analyze a new type of stochastic first order method: gradient descent with compressed iterates (GDCI). GDCI in each iteration first compresses the current iterate using a lossy randomized compression technique, and subsequently takes a gradient step. This method is a distillation of a key ingredient in the current practice of federated learning, where a model needs to be compressed by a mobile device before it is sent back to a server for aggregation. Our analysis provides a step towards closing the gap between the theory and practice of federated learning, and opens the possibility for many extensions.

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。

↑