arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1806.07572cs.LGcs.NEmath.PRstat.ML
收藏

神经切线核:神经网络的收敛性和推广性

Neural Tangent Kernel: Convergence and Generalization in Neural Networks

École Polytechnique Fédérale de Lausanne(洛桑联邦理工学院);Imperial College London(帝国理工学院)

Arthur Jacot, Franck Gabriel, Clément Hongler

摘要:

在初始化时,人工神经网络(ANN)等价于无限宽度限制下的高斯过程,从而将它们连接到核方法。我们证明了人工神经网络在训练过程中的演变也可以描述的内核:在梯度下降的人工神经网络的参数,网络功能$f_θ$(将输入向量映射到输出向量)遵循内核梯度的功能成本(这是凸的,在对比参数成本)w.r.t.一个新的内核:神经切线核(NTK)。该核是描述人工神经网络泛化特征的核心。虽然NTK在初始化时是随机的,并且在训练过程中变化,但在无限宽度限制下,它收敛到一个显式限制核,并且在训练过程中保持不变。这使得在函数空间而不是参数空间中研究人工神经网络的训练成为可能。然后,训练的收敛性可以与限制NTK的正定性相关。证明了当数据支撑在球面上且非线性为非多项式时,极限NTK的正定性。然后,我们专注于最小二乘回归的设置,并表明在无限宽的限制,网络函数$f_θ$遵循线性微分方程在训练过程中。相对于NTK,沿着输入数据的最大核主成分的收敛是最快的,因此提出了提前停止的理论动机。最后,我们研究了NTK数值,观察其行为的宽网络,并将其与无限宽的限制。

英文摘要:

At initialization, artificial neural networks (ANNs) are equivalent to Gaussian processes in the infinite-width limit, thus connecting them to kernel methods. We prove that the evolution of an ANN during training can also be described by a kernel: during gradient descent on the parameters of an ANN, the network function $f_θ$ (which maps input vectors to output vectors) follows the kernel gradient of the functional cost (which is convex, in contrast to the parameter cost) w.r.t. a new kernel: the Neural Tangent Kernel (NTK). This kernel is central to describe the generalization features of ANNs. While the NTK is random at initialization and varies during training, in the infinite-width limit it converges to an explicit limiting kernel and it stays constant during training. This makes it possible to study the training of ANNs in function space instead of parameter space. Convergence of the training can then be related to the positive-definiteness of the limiting NTK. We prove the positive-definiteness of the limiting NTK when the data is supported on the sphere and the non-linearity is non-polynomial. We then focus on the setting of least-squares regression and show that in the infinite-width limit, the network function $f_θ$ follows a linear differential equation during training. The convergence is fastest along the largest kernel principal components of the input data with respect to the NTK, hence suggesting a theoretical motivation for early stopping. Finally we study the NTK numerically, observe its behavior for wide networks, and compare it to the infinite-width limit.

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。

↑