Learning Brave Assumption-Based Argumentation Frameworks via ASP

TL;DR

This paper introduces a novel method for automatically learning Assumption-based Argumentation frameworks from data using Answer Set Programming, enhancing the automation of non-monotonic reasoning models.

Contribution

It presents a new algorithm for learning ABA frameworks from examples, framing the problem as brave reasoning, and implements it with ASP, outperforming existing ILP systems.

Findings

The proposed algorithm effectively learns ABA frameworks from data.

The ASP-based implementation demonstrates practical efficiency.

Comparison shows improvements over state-of-the-art ILP systems.

Abstract

Assumption-based Argumentation (ABA) is advocated as a unifying formalism for various forms of non-monotonic reasoning, including logic programming. It allows capturing defeasible knowledge, subject to argumentative debate. While, in much existing work, ABA frameworks are given up-front, in this paper we focus on the problem of automating their learning from background knowledge and positive/negative examples. Unlike prior work, we newly frame the problem in terms of brave reasoning under stable extensions for ABA. We present a novel algorithm based on transformation rules (such as Rote Learning, Folding, Assumption Introduction and Fact Subsumption) and an implementation thereof that makes use of Answer Set Programming. Finally, we compare our technique to state-of-the-art ILP systems that learn defeasible knowledge.