What is it about?

DLV2 is an AI tool for Knowledge Representation and Reasoning which supports Answer Set Programming (ASP) – a logic-based declarative formalism, successfully used in both academic and industrial applications. This work presents a new incremental reasoner obtained from the evolution of DLV2 towards multi-shot reasoning. Rather than restarting the computation from scratch, the system remains alive and incrementally handles the internal grounding process: in a completely transparent fashion for the user, at each shot, it reuses previous computations for building and maintaining a large, more general ground program, from which a smaller yet equivalent portion is determined and used for computing answer sets. The paper describes the system, its usage, its applicability and performance in some practically relevant domains.

Featured Image

Why is it important?

Incremental-DLV2 is a new incremental ASP reasoner that represents the evolution of DLV2 towards multi-shot reasoning. DLV2 is a novel version of one of the first and more widespread ASP system, namely DLV; it has been re-implemented from scratch and encompasses the outcome of the latest research effort on both grounding and solving areas. Just as DLV and DLV2, Incremental-DLV2 fully embraces the declarative nature of ASP and transparently handles incrementality via overgrounding. This makes, at each shot, the instantiation effort directly proportional to the number of unseen facts, up to the point that the effort is null when all input facts have been already seen in previous shots. Moreover, since producing a larger and larger ground program could negatively impact on the solving step, Incremental-DLV2 properly selects only a smaller yet equivalent portion to be considered during solving. The approach paves the way to the use of ASP in real-world contexts where repeated reasoning tasks have to be performed, such as robot controlling systems, and is stream reasoning scenarios, such as smart cities.

Perspectives

In the latest years, emerging application contexts, such as realtime motion tracking, content distribution, robotics, artificial players in videogames, sensor network configuration, have been posing new challenges for Intelligent Systems that are required to show high reactivity while performing the repeated execution of reasoning tasks over rapidly changing input facts. The work herein presented opens several interesting research perspectives: it paves the way to the use of Answer Set Programming in such real-world contexts, thus enabling the successful development of applications based on Declarative and Explainable Artificial Intelligence.

Simona Perri
Universita degli Studi della Calabria

The work herein presented paves the way to a more successful use of Declarative and Explainable Artificial Intelligence in real world contexts.

Francesco Calimeri
Universita degli Studi della Calabria

Read the Original

This page is a summary of: ASP-based Multi-shot Reasoning via DLV2 with Incremental Grounding, September 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3551357.3551371.
You can read the full text:

Read

Contributors

The following have contributed to this page