Toeplitz And Circulant Matrices: A Review (Foundations and Trends(R) in Communications and Information Theory)

461 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Distinguishing Cause from Effect Using Observational Data: Methods and Benchmarks (2014, background): “As the product of two Gaussian kernels is characteristic, HSIC with such kernels will detect any dependence asymptotically (see also Lemma 12 in Appendix A), at least when the bandwidths are fixed.”
The Singular Values of Convolutional Layers (2018, background): “The quality of this approximation has been heavily analyzed in the case of one-dimensional signals (Gray, 2006).”
Circulant Binary Embedding (2014, unknown): “A circulant matrixR ∈ Rd×d is a matrix defined by a vector r = (r0, r2, · · · , rd−1)T (Gray, 2006)4.”
Kaleidoscope: An Efficient, Learnable Representation For All Structured Linear Maps (2020, background): “…discrete cosine transform (DCT), discrete sine transform (DST), and Hadamard transform), convolution (i.e. circulant matrices), Toeplitz matrices (Gray, 2006), structured matrices for kernel approximation ( ( HD ) 3 (Yu et al., 2016)) and compact neural network design (Fastfood (Le et al.,…”
Entropy and Kullback-Leibler divergence estimation based on Szegö's theorem (2009, background): “A useful property, that will be used later, is that the Kullback-Leibler divergence is scale invariant.”