Program checkers for probability generation
[For the entire collection see Zbl 0753.00027.] ``Program checking was introduced by \textit{M. Blum} and \textit{S. Kannan} [Designing Programs That Check Their Work, Proc. 21st ACM Symposium on Theory of Computing, 86-97 (1989)] as a way of certifying the correctness of program output. The original definition applied only to computational problems where correctness could be described as a relation between input and output. The idea was to ensure the correctness of the output of a program (with high probability) on each of its runs rather than on proving the program correct overall. In the process of checking the program on any one run the checker was allowed to make additional calls to the program and was allowed to declare the program buggy if the program produced wrong output on any of the calls. We are interested in extending the concept of program checking to programs that claim to generate objects according to probabilistic distributions.
- Efficient checkers for number-theoretic computations
- scientific article; zbMATH DE number 4195192 (Why is no real title available?)
- Designing programs that check their work
- scientific article; zbMATH DE number 1919505 (Why is no real title available?)
- Program result checking: a new approach to making programs more reliable
- Does a Program Yield the Right Distribution?
- Pipelined algorithms to detect cheating in long-term grid computations
This page was built for publication: Program checkers for probability generation
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4036561)
