Papers
Topics
Authors
Recent
Search
2000 character limit reached

AFRA: Argumentation framework with recursive attacks

Published 11 Oct 2018 in cs.AI | (1810.04886v1)

Abstract: The issue of representing attacks to attacks in argumentation is receiving an increasing attention as a useful conceptual modelling tool in several contexts. In this paper we present AFRA, a formalism encompassing unlimited recursive attacks within argumentation frameworks. AFRA satisfies the basic requirements of definition simplicity and rigorous compatibility with Dung's theory of argumentation. This paper provides a complete development of the AFRA formalism complemented by illustrative examples and a detailed comparison with other recursive attack formalizations.

Citations (165)

Summary

Overview of AFRA: Argumentation Framework with Recursive Attacks

The paper presents AFRA: Argumentation Framework with Recursive Attacks, aiming to extend Dung's seminal concept of argumentation frameworks (AF) by enabling recursive attacks on attacks. Dung's framework is foundational in argumentation theory, enabling abstract representation and analysis of argument relations, notably the attack relation. AFRA seeks to generalize this framework, preserving the fundamental structure while providing greater flexibility for complex reasoning patterns.

Key Features and Definitions

AFRA augments AF by allowing attacks not only against arguments but also against other attacks. The authors introduce a series of definitions essential for understanding AFRA’s structure, including conflict-freeness, admissibility, and the novel notions of direct and indirect defeat. Central to AFRA is its Algorithm for Argumentation Reasoning, which ensures the compatibility of recursive attack semantics with Dung's established theories.

The recursive attack capability in AFRA allows it to represent nuanced argumentation scenarios that are challenging to capture using Dung’s original framework. An illustrative example involves decision-making contexts, such as evaluating recursive preferences and counter-preferences when deciding between holiday destinations under various conditions. This example emphasizes AFRA's power to encapsulate argument dynamics where attacks and counterattacks do not merely resolve to direct hypocritical statements but extend to contextual and relational complexities.

Comparisons and Compatibility

The paper provides substantive comparisons between AFRA and related formalisms such as Extended Argumentation Framework (EAF) and Higher Order Argumentation Framework (HOAF). EAF, a precursor marked by a limitation of one-level recursion in attack, falls short in areas AFRA seeks to enrich. HOAF, on the other hand, uses meta-level arguments to express attacks, offering AFRA’s simpler syntactic model as a beneficial complement.

Compatibility with traditional AF is a cornerstone discussed by the authors, emphasizing bijective correspondences between AFRA semantics and Dung’s semantics. The established relations affirm AFRA’s allegiance to Dung’s principles, thus maintaining meaningful links to exhaustively studied argumentation semantics like grounded, preferred, and stable extensions.

Implications

The recursive nature of AFRA opens the door to formalizing intricate argumentation problems that were previously difficult to manage. Recursive attacks allow natural modeling of preference dynamics, coalition reasoning, and meta-argumentation contexts, yielding implications for fields like artificial intelligence, legal reasoning, and decision support systems.

From a computational perspective, the transitions and transformations provided between AFRA and traditional AF semantics suggest that many known computational complexity results from AF can be transferred and applied within AFRA frameworks. This potential transferability is crucial for the future implementation and scalability of argumentation systems that adopt this more expressive framework.

Directions for Future Research

Though AFRA presents a substantial advancement, several avenues warrant exploration. Detailed complexity analyses, especially concerning recursive reasoning, could significantly enhance understanding and implementation. Furthermore, AFRA’s integration with existing AI systems and reasoning software remains an open frontier, with tools like ASPARTIX demonstrating the practical feasibility of recursive attack deployment.

In conclusion, the development of AFRA positions it as an impactful contribution to the field of argumentation theory, offering both theoretical insights and practical utilities for complex reasoning landscapes. Its ability to incorporate recursive attacks effectively enriches the expressive power and flexibility of argumentation models, advancing the broader quest for computational systems that cogently represent human-like argumentation.

Paper to Video (Beta)

No one has generated a video about this paper yet.

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Continue Learning

We haven't generated follow-up questions for this paper yet.

Collections

Sign up for free to add this paper to one or more collections.