Abstract

We present a method for making use of past proof experience called flexible re-enactment (FR). FR is actually a search-guiding heuristic that uses past proof experience to create a search bias. Given a proof P of a problem solved previously that is assumed to be similar to the current problem A, FR searches for P and in the "neighborhood" of P in order to find a proof of A. This heuristic use of past experience has certain advantages that make FR quite profitable and give it a wide range of applicability. Experimental studies substantiate and illustrate this claim. This work was supported by the Deutsche Forschungsgemeinschaft (DFG). 2 1 INTRODUCTION 1 Introduction Automated deduction is essentially a search problem that gives rise to potentially infinite search spaces because of general undecidability. Despite these unfavorable conditions state-of-the-art theorem provers have gained a remarkable level of performance mainly due to (problem-specific) search-guiding heuristics and...

Citations