arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1502.06800cs.LGcs.NAmath.NAstat.ML
收藏

核积分规则与随机特征展开的等价性

On the Equivalence between Kernel Quadrature Rules and Random Feature Expansions

INRIA(法国国家信息与自动化研究所)

Francis Bach

AI总结:

研究揭示核积分规则是随机特征展开的特例,通过理论分析得出样本数与积分算子特征值的关系,扩展至函数逼近问题并改进随机特征学习的泛化保证。

摘要:

我们展示基于核的积分规则可以视为正定核随机特征展开的特例,对于特定分解总存在。我们提供理论分析,得出所需样本数与近似误差的关系,得到基于积分算子特征值的上下界,匹配对数项。特别地,我们显示上界可通过特定非均匀分布的独立同分布样本获得,而下界若对任何点集有效。将结果应用于核积分规则时,我们恢复了Sobolev空间的已知上下界。此外,结果扩展至更一般的函数逼近问题,得到L2-和L∞-范数结果,匹配特殊情形的已知结果。应用于随机特征时,我们显示改进了保持学习Lipschitz连续损失泛化保证所需的随机特征数量。

英文摘要:

We show that kernel-based quadrature rules for computing integrals can be seen as a special case of random feature expansions for positive definite kernels, for a particular decomposition that always exists for such kernels. We provide a theoretical analysis of the number of required samples for a given approximation error, leading to both upper and lower bounds that are based solely on the eigenvalues of the associated integral operator and match up to logarithmic terms. In particular, we show that the upper bound may be obtained from independent and identically distributed samples from a specific non-uniform distribution, while the lower bound if valid for any set of points. Applying our results to kernel-based quadrature, while our results are fairly general, we recover known upper and lower bounds for the special cases of Sobolev spaces. Moreover, our results extend to the more general problem of full function approximations (beyond simply computing an integral), with results in L2- and L$\infty$-norm that match known results for special cases. Applying our results to random features, we show an improvement of the number of random features needed to preserve the generalization guarantees for learning with Lipschitz-continuous losses.

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。

↑