![]()
Type-Theoretic Characterizations of the Basic Feasible FunctionalsJames S. Royer |
|---|
![]()
The usual way of describing a bound on the time or space complexity of an algorithm is to give some expression, e.g., a polynomial, over the sizes of arguments of the algorithm. When the algorithm involves higher-type arguments, however, there is no commonly accepted notion of the "size" of a higher-type object nor, for that matter, of a "higher-type polynomial."
For type-level two, Kapron and Cook have proposed a notion of the length of a type-one object and of a notion of second-order polynomial. Moreover, they showed that, with respect to a reasonable machine model, the type-two "polynomial-time" computable functionals are exactly the type-2 basic feasible functionals (BFF2), where BFF2 is a class of functionals first described (via a programming formalism) in 1974 by Mehlhorn and later, independently, by Cook and Urquhart in 1989 (via a different programming formalism).
While this a deep and difficult characterization, it does not leave matters in a complete satisfactory state. First, on the programming formalism side, all the the prior programming formalism for BFF2 are difficult to actually write programs in. Second, on the machine/poly-bounds side, programming is much easier in this context, but there is an air of ad hoc-ery the machine model and the poly-bounds.
In this talk we will present a characterization of BFF2 based on type-theoretic constraints inspired by type-theoretic characterizations of type-1 polynomial-time by Bellantoni and Cook and by Leivant. While the Bellantoni-Cook and Leivant systems are strictly predicative, it seems necessary to partially break predicativity under our general approach. Also, in contrast to the prior programming formalisms characterizing BFF2, our formalism is very close in spirit to Kapron and Cook's machine-based characterization of BFF2.
If there is time, we will also discuss
This is joint work with Robert Irwin and Bruce Kapron.
![]()