On the Configuration of More and Less Expressive Logic Programs
This work addresses the issue of solver performance variability for AI practitioners, but it is incremental as it builds on existing configuration methods.
The paper tackled the problem of improving reasoning in model-based AI by exploiting solver sensitivity to syntactic changes, using automated configuration tools to reformulate inputs for SAT and ASP. Results from experiments on competition domains showed various advantages from this input reformulation approach.
The decoupling between the representation of a certain problem, i.e., its knowledge model, and the reasoning side is one of main strong points of model-based Artificial Intelligence (AI). This allows, e.g. to focus on improving the reasoning side by having advantages on the whole solving process. Further, it is also well-known that many solvers are very sensitive to even syntactic changes in the input. In this paper, we focus on improving the reasoning side by taking advantages of such sensitivity. We consider two well-known model-based AI methodologies, SAT and ASP, define a number of syntactic features that may characterise their inputs, and use automated configuration tools to reformulate the input formula or program. Results of a wide experimental analysis involving SAT and ASP domains, taken from respective competitions, show the different advantages that can be obtained by using input reformulation and configuration. Under consideration in Theory and Practice of Logic Programming (TPLP).