Timezone: »

 
Poster
Gaussian Quadrature for Kernel Features
Tri Dao · Christopher M De Sa · Christopher Ré

Tue Dec 05 06:30 PM -- 10:30 PM (PST) @ Pacific Ballroom #56
in Posters Tue »
Kernel methods have recently attracted resurgent interest, showing performance competitive with deep neural networks in tasks such as speech recognition. The random Fourier features map is a technique commonly used to scale up kernel machines, but employing the randomized feature map means that $O(\epsilon^{-2})$ samples are required to achieve an approximation error of at most $\epsilon$. We investigate some alternative schemes for constructing feature maps that are deterministic, rather than random, by approximating the kernel in the frequency domain using Gaussian quadrature. We show that deterministic feature maps can be constructed, for any $\gamma > 0$, to achieve error $\epsilon$ with $O(e^{\gamma} + \epsilon^{-1/\gamma})$ samples as $\epsilon$ goes to 0. Our method works particularly well with sparse ANOVA kernels, which are inspired by the convolutional layer of CNNs. We validate our methods on datasets in different domains, such as MNIST and TIMIT, showing that deterministic features are faster to generate and achieve accuracy comparable to the state-of-the-art kernel methods based on random Fourier features.
Keywords: Kernel Methods · Audio and Speech Processing

Author Information

Tri Dao (Stanford University)
Christopher M De Sa (Stanford)
Christopher Ré (Stanford)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors