Title :
k-Independent Gaussians Fool Polynomial Threshold Functions
Author_Institution :
Dept. of Math., Harvard Univ., Cambridge, MA, USA
Abstract :
We show that limited independence suffices to fool polynomial threshold functions of arbitrary degree.
Keywords :
Gaussian processes; computational complexity; arbitrary degree; independence suffices; k-independent Gaussians fool polynomial threshold functions; Approximation methods; Correlation; Logic functions; Polynomials; Random variables; Silicon; Taylor series; Gaussian distributions; Polynomials; Threshold logic functions;
Conference_Titel :
Computational Complexity (CCC), 2011 IEEE 26th Annual Conference on
Conference_Location :
San Jose, CA
Print_ISBN :
978-1-4577-0179-5
Electronic_ISBN :
1093-0159
DOI :
10.1109/CCC.2011.13
