arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.11558cs.LG
收藏

加权神经切线核:一种推广和改进的网络诱导核

Weighted Neural Tangent Kernel: A Generalized and Improved Network-Induced Kernel

Shanghai Jiao Tong University(上海交通大学)

Lei Tan, Shutong Wu, Xiaolin Huang

摘要:

神经切线核(NTK)最近吸引了大量的研究,因为它描述了由梯度下降训练的过参数化神经网络(NN)的演化。然而,现在众所周知的是,梯度下降并不总是一个很好的优化神经网络,这可以部分解释NTK回归估计器的不令人满意的实际性能。在本文中,我们介绍了加权神经切线核(WNTK),一个广义的和改进的工具,它可以捕捉一个过参数化的神经网络的训练动态下,不同的优化。在理论上,在无限宽的限制下,我们证明了:i)WNTK在初始化和训练过程中的稳定性,以及ii)WNTK回归估计量和相应的NN估计量之间的等价性,对于不同的参数具有不同的学习率。在数值实验中,经验和解析WNTK都优于相应的NTK。

英文摘要:

The Neural Tangent Kernel (NTK) has recently attracted intense study, as it describes the evolution of an over-parameterized Neural Network (NN) trained by gradient descent. However, it is now well-known that gradient descent is not always a good optimizer for NNs, which can partially explain the unsatisfactory practical performance of the NTK regression estimator. In this paper, we introduce the Weighted Neural Tangent Kernel (WNTK), a generalized and improved tool, which can capture an over-parameterized NN's training dynamics under different optimizers. Theoretically, in the infinite-width limit, we prove: i) the stability of the WNTK at initialization and during training, and ii) the equivalence between the WNTK regression estimator and the corresponding NN estimator with different learning rates on different parameters. With the proposed weight update algorithm, both empirical and analytical WNTKs outperform the corresponding NTKs in numerical experiments.

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。

↑