By Topic

The average amount of information lost in multiplication

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

1 Author(s)
Pippenger, N. ; Dept. of Comput. Sci., Princeton Univ., NJ

We show that if X and Y are integers independently and uniformly distributed in the set {1,...,N}, then the information lost in forming their product (which is given by the equivocation H(X,Y|XmiddotY)), is Theta(loglogN). We also prove two extremal results regarding cases in which X and Y are not necessarily independently or uniformly distributed. First, we note that the information lost in multiplication can of course be 0. We show that the condition H(X,Y|XmiddotY)=0 implies 2log2N-H(X,Y)=Omega(loglogN). Furthermore, if X and Y are independent and uniformly distributed on disjoint sets of primes, it is possible to have H(X,Y|XmiddotY)=0 with log2N-H(X) and log2N-H(Y) each O(loglogN). Second, we show that no matter how X and Y are distributed, H(X,Y|XmiddotY)=O(logN/loglogN). Furthermore, there are distributions (in which X and Y are independent and uniformly distributed over sets of numbers having only small and distinct prime factors) for which we have H(X,Y|XmiddotY)=Omega(logN/loglogN)

Published in:

Information Theory, IEEE Transactions on  (Volume:51 ,  Issue: 2 )