Bias

A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z

Definition(s):

With respect to the uniform distribution on \([0,n-1]\), the bias is defined to be the maximum value of \(\{probability(S) - (\frac{|S|}{n})\}\) taken over all subsets \(S\) of \([0,n-1]\). This measures the maximum advantage that an adversary has in predicting any event.
Source(s):
FIPS 186-5

Glossary Comments

Comments about specific definitions should be sent to the authors of the linked Source publication. For NIST publications, an email is usually found within the document.

Comments about the glossary's presentation and functionality should be sent to secglossary@nist.gov.

See NISTIR 7298 Rev. 3 for additional details.