ABSTRACT
We give an introduction to the design of parallel algorithms with the probabilistic method. Algorithms of this kind usually possess a randomized sequential counterpart. Parallelization of such algorithms is inherently linked with derandomization, either with the Erdo˝s-Spencer method of conditional probabilities or exhaustive search in a polynomial-sized sample space.
