Logistic AI
Is a traditional field within AI that hopes to build on programs of logic (theorem provers, deductive algorithms, etc.) to create intelligent systems.
The main obstacles to this approach are:
- It's not easy to take informal knowledge and formulate it in terms required by logical notation (particularly when the knowledge isn't 100% certain).
- There is a big difference between solving a problem in principle and in practice. Even a problem with a few hundred facts could exhaust a computer resources and require guidance to solve it in an acceptable amount of time.