The notion of efficient proof-checking has always been central to complexit
y theory, and it gave rise to the definition of the class NP. In the last 1
5 years there has been a number of exciting, unexpected and deep developmen
ts in complexity theory that exploited the notion of randomized and interac
tive proof-checking. Results developed along this line of research have div
erse and powerful applications in complexity theory, cryptography, and the
theory of approximation algorithms for combinatorial optimization problems.
In this paper we survey the main lines of developments in interactive and
probabilistic proof-checking, with an emphasis on open questions. (C) 2000
Elsevier Science B.V. All rights reserved. MSC. 68Q10; 68Q15; 68Q17; 03F20.