Attack Game

Updated 11 Aug 2026

•
What it is
- •
  the book's core formal methodology for defining security: a protocol between a challenger (follows a fixed, simple protocol) and an adversary (follows an arbitrary efficient protocol), often split into two experiments (Experiment 0 / Experiment 1) that differ only in what the challenger does
•
Advantage
- •
  for a security property "X," the adversary's advantage is $\mathrm{Xadv}[A, S] := |\Pr[W_0] - \Pr[W_1]|$ , where $W_b$ is the event the adversary "wins" (outputs 1) in Experiment $b$
- •
  a system is considered secure (with respect to X) if every efficient adversary has negligible advantage
•
Used throughout the book
- •
  this same challenger/adversary/two-experiment pattern defines nearly every security notion in the book (semantic security, message recovery, chosen plaintext/ciphertext security, unforgeability, etc.) — learning this pattern once pays off repeatedly
•
Note
- •
  the Coursera lecture version of this same idea is framed as a Statistical Test with a "distinguishing advantage" — same underlying concept, different course's vocabulary
•
Related

Linked references 6