Nearly optimal deterministic algorithm for sparse Walsh-Hadamard transform
For every fixed constant α > 0, we design an algorithm for computing the k-sparse Walsh-Hadamard transform (i.e., Discrete Fourier Transform over the Boolean cube) of an N-dimensional vector x ∈ R[superscript N] in time k[superscript 1 + α](log N)[superscript O(1)]. Specifically, the algorithm is...
Main Authors: | , |
---|---|
Other Authors: | , |
Format: | Article |
Language: | English |
Published: |
Association for Computing Machinery,
2018-02-26T21:44:54Z.
|
Subjects: | |
Online Access: | Get fulltext |