Title: Comparing Computational Power - Part II

Udi Boker

In this talk Udi Boker will get into details not covered in his previous talk on the subject on January 18th. In particular, he will compare the recursive functions, Turing machines and the lambda calculus from the point of view of their robustness.

Abstract of his previous talk: It is standard to assert that a computational model (set of partial functions) M is as powerful as a model M' by fixing a 1-1 mapping between their domains and showing that every function of M' is simulated by some function of M. We show that in the general case a model can be of equivalent power to a proper superset of itself. Define a model that is strictly weaker than all its supersets as 'robust'. We describe several results regarding robustness. In particular, the set of recursive functions is robust. Accordingly, we provide a criterion for verifying that a model, operating over an enumerable domain, is super-recursive.