Counting with a fixed supplied family #
For any fixed supplied : U^n → U^k, the number of targets with conditional
gate complexity at most budget is bounded by the number of circuit
descriptions on n + k inputs. The bound is independent of the complexity
of supplied. Dividing by the size of a nonempty target family gives a bound
on the probability that a uniformly sampled target is conditionally easy.
The supplied family is fixed before choosing the target. In particular,
these statements do not bound an adversarial choice supplied = target.
All targets computable within a gate budget from a fixed supplied family.
Equations
- One or more equations did not get rendered due to their size.
Instances For
Membership in the counted set is exactly a conditional complexity bound.
Fixed supplied functions create at most one target per circuit description.
The factorial-improved semantic count also bounds conditional complexity.
Any target family larger than the circuit budget contains a function
that remains hard after supplying supplied.
Factorial-improved counting yields a conditionally hard target.
Counting over the entire truth-table space yields a conditionally hard target.
A finite menu of supplied families costs only a multiplicative factor in counting. A sufficiently large target family has a member hard for every choice from the menu, even if that choice is made after seeing the target.
Probability bound for a uniformly sampled member of a nonempty target family, expressed as an exact rational cardinality ratio. The supplied family is fixed independently of this sampling.