通过链式方法改进核密度估计器的更强核心集界
Stronger Coreset Bounds for Kernel Density Estimators via Chaining
AI总结:
本文利用差异方法和链式方法,为高斯、拉普拉斯、指数、Hellinger和JS核函数改进了核心集复杂度界,并给出随机多项式时间算法生成更小核心集。
摘要:
我们应用差异方法和链式方法,为一大类核函数的核心集复杂度提供了改进的界。我们的结果给出了随机多项式时间算法,在数据集一致有界的情况下,为高斯核和拉普拉斯核生成大小为 $O\big(\frac{\sqrt{d}}{\varepsilon}\sqrt{\log\log \frac{1}{\varepsilon}}\big)$ 的核心集,这是以往技术无法实现的改进。对于常数 $d$,我们还为拉普拉斯核获得了大小为 $O\big(\frac{1}{\varepsilon}\sqrt{\log\log \frac{1}{\varepsilon}}\big)$ 的核心集。最后,我们给出了指数核、Hellinger核和JS核的核心集复杂度的最佳已知界 $O\big(\frac{\sqrt{d}}{\varepsilon}\sqrt{\log(2\max\{1,\alpha\})}\big)$,其中 $1/\alpha$ 是核的带宽参数。
英文摘要:
We apply the discrepancy method and a chaining approach to give improved bounds on the coreset complexity of a wide class of kernel functions. Our results give randomized polynomial time algorithms to produce coresets of size $O\big(\frac{\sqrt{d}}{\varepsilon}\sqrt{\log\log \frac{1}{\varepsilon}}\big)$ for the Gaussian and Laplacian kernels in the case that the data set is uniformly bounded, an improvement that was not possible with previous techniques. We also obtain coresets of size $O\big(\frac{1}{\varepsilon}\sqrt{\log\log \frac{1}{\varepsilon}}\big)$ for the Laplacian kernel for $d$ constant. Finally, we give the best known bounds of $O\big(\frac{\sqrt{d}}{\varepsilon}\sqrt{\log(2\max\{1,α\})}\big)$ on the coreset complexity of the exponential, Hellinger, and JS Kernels, where $1/α$ is the bandwidth parameter of the kernel.
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。