Universality in numerical computations with random data.

Proc Natl Acad Sci U S A

Courant Institute, New York University, New York, NY 10012;

Published: October 2014

The authors present evidence for universality in numerical computations with random data. Given a (possibly stochastic) numerical algorithm with random input data, the time (or number of iterations) to convergence (within a given tolerance) is a random variable, called the halting time. Two-component universality is observed for the fluctuations of the halting time--i.e., the histogram for the halting times, centered by the sample average and scaled by the sample variance, collapses to a universal curve, independent of the input data distribution, as the dimension increases. Thus, up to two components--the sample average and the sample variance--the statistics for the halting time are universally prescribed. The case studies include six standard numerical algorithms as well as a model of neural computation and decision-making. A link to relevant software is provided for readers who would like to do computations of their own.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4210305PMC
http://dx.doi.org/10.1073/pnas.1413446111DOI Listing

Publication Analysis

Top Keywords

universality numerical
8
numerical computations
8
computations random
8
random data
8
input data
8
halting time
8
sample average
8
random
4
data
4
data authors
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!