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 2012.07205math.NAcs.NA
收藏

具有余弦和ReLU $^k$激活函数的浅神经网络的高阶逼近率

High-Order Approximation Rates for Shallow Neural Networks with Cosine and ReLU$^k$ Activation Functions

Jonathan W. Siegel, Jinchao Xu

摘要:

研究了激活函数为线性单元的幂的浅层神经网络的逼近性质。具体来说,我们认为依赖的逼近速度的尺寸和光滑性的基础函数$f$的谱巴伦空间被近似。我们发现,随着$f$的平滑指数$s$的增加,具有ReLU $^k$激活函数的浅层神经网络获得了改进的逼近率,最高可达$L^2$中$O(n^{-(k+1)}\log(n))$的最佳可能速率,与维度$d$无关。该结果的意义在于,激活函数ReLU $^k$是固定的,与维数无关,而对于经典方法,为了利用$f$的维数相关的平滑性,所使用的小波的多项式逼近度或平滑度必须增加。此外,我们得到改进的近似率的浅神经网络与余弦激活函数的谱巴伦空间。最后,我们证明了下界表明,在给定的假设下,所获得的逼近率是最优的。

英文摘要:

We study the approximation properties of shallow neural networks with an activation function which is a power of the rectified linear unit. Specifically, we consider the dependence of the approximation rate on the dimension and the smoothness in the spectral Barron space of the underlying function $f$ to be approximated. We show that as the smoothness index $s$ of $f$ increases, shallow neural networks with ReLU$^k$ activation function obtain an improved approximation rate up to a best possible rate of $O(n^{-(k+1)}\log(n))$ in $L^2$, independent of the dimension $d$. The significance of this result is that the activation function ReLU$^k$ is fixed independent of the dimension, while for classical methods the degree of polynomial approximation or the smoothness of the wavelets used would have to increase in order to take advantage of the dimension dependent smoothness of $f$. In addition, we derive improved approximation rates for shallow neural networks with cosine activation function on the spectral Barron space. Finally, we prove lower bounds showing that the approximation rates attained are optimal under the given assumptions.

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。

↑