Sigma-Delta and Distributed Noise-Shaping Quantization Methods for Random Fourier Features
From MaRDI portal
Random fields; image analysis (62M40) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Computing methodologies for image processing (68U10) Numerical aspects of computer graphics, image analysis, and computational geometry (65D18) Numerical methods for discrete and fast Fourier transforms (65T50)
Abstract: We propose the use of low bit-depth Sigma-Delta and distributed noise-shaping methods for quantizing the Random Fourier features (RFFs) associated with shift-invariant kernels. We prove that our quantized RFFs -- even in the case of -bit quantization -- allow a high accuracy approximation of the underlying kernels, and the approximation error decays at least polynomially fast as the dimension of the RFFs increases. We also show that the quantized RFFs can be further compressed, yielding an excellent trade-off between memory use and accuracy. Namely, the approximation error now decays exponentially as a function of the bits used. Moreover, we empirically show by testing the performance of our methods on several machine learning tasks that our method compares favorably to other state of the art quantization methods in this context.
Recommendations
- Breaking the waves: asymmetric random periodic features for low-bitrate kernel machines
- Quasi-Monte Carlo feature maps for shift-invariant kernels
- Towards a unified analysis of random Fourier features
- scientific article; zbMATH DE number 7255176
- Distributed noise-shaping quantization. I: Beta duals of finite frames and near-optimal quantization of random measurements
This page was built for publication: Sigma-Delta and Distributed Noise-Shaping Quantization Methods for Random Fourier Features
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6369474)
