arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.10741cs.LGmath.OCstat.ML
收藏

自然梯度的迹限制Kronecker因子逼近

A Trace-restricted Kronecker-Factored Approximation to Natural Gradient

Tianjin University(天津大学);Huawei Technologies(华为技术有限公司);Beijing University of Technology(北京工业大学)

Kai-Xin Gao, Xiao-Lei Liu, Zheng-Hai Huang, Min Wang, Zidong Wang, Dachuan Xu, Fan Yu

摘要:

二阶优化方法具有通过曲率矩阵修改梯度来加速收敛的能力。已经有许多尝试使用二阶优化方法来训练深度神经网络。受对角逼近和Kronecker-Factored Approximate Curvature(KFAC)等因子逼近方法的启发,提出了一种新的Fisher信息矩阵(Fisher information matrix)的迹限制Kronecker-Factored Approximate Curvature(TKFAC)逼近方法,该方法能够保持精确信息矩阵和近似信息矩阵之间的迹关系.在TKFAC中,我们将近似矩阵的每个块分解为两个较小矩阵的Kronecker积,并通过与迹相关的系数进行缩放。从理论上分析了TKFAC的逼近误差,给出了逼近误差的上界,并提出了一种新的阻尼技术,用于卷积神经网络上的TKFAC,以保持二阶优化方法在训练过程中的优势。实验表明,我们的方法具有更好的性能相比,一些国家的最先进的算法在一些深度网络架构。

英文摘要:

Second-order optimization methods have the ability to accelerate convergence by modifying the gradient through the curvature matrix. There have been many attempts to use second-order optimization methods for training deep neural networks. Inspired by diagonal approximations and factored approximations such as Kronecker-Factored Approximate Curvature (KFAC), we propose a new approximation to the Fisher information matrix (FIM) called Trace-restricted Kronecker-factored Approximate Curvature (TKFAC) in this work, which can hold the certain trace relationship between the exact and the approximate FIM. In TKFAC, we decompose each block of the approximate FIM as a Kronecker product of two smaller matrices and scaled by a coefficient related to trace. We theoretically analyze TKFAC's approximation error and give an upper bound of it. We also propose a new damping technique for TKFAC on convolutional neural networks to maintain the superiority of second-order optimization methods during training. Experiments show that our method has better performance compared with several state-of-the-art algorithms on some deep network architectures.

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。

↑