Search

Explainable Artificial Intelligence (XAI) is a research area that clarifies AI decision-making processes to build user trust and promote responsible AI. Hence, a key scientific challenge in XAI is the development of methods that generate transparent and interpretable explanations while maintaining scalability and effectiveness in complex scenarios. Rule-based methods in XAI generate rules that can potentially explain AI inferences, yet they can also become convoluted in large scenarios, hindering their readability and scalability. Moreover, they often lack contrastive explanations, leaving users uncertain why specific predictions are preferred. To address this scientific problem, we explore the integration of computational argumentation—a sub-field of AI that models reasoning processes through defeasibility—into rule-based XAI systems. Computational argumentation enables arguments modelled from rules to be retracted based on new evidence. This makes it a promising approach to enhancing rule-based methods for creating more explainable AI systems. Nonetheless, research on their integration remains limited despite the appealing properties of rule-based systems and computational argumentation. Therefore, this study also addresses the applied challenge of implementing such an integration within practical AI tools. The study employs the Logic Learning Machine (LLM), a specific rule-extraction technique, and presents a modular design that integrates input rules into a structured argumentation framework using state-of-the-art computational argumentation methods. Experiments conducted on binary classification problems using various datasets from the UCI Machine Learning Repository demonstrate the effectiveness of this integration. The LLM technique excelled in producing a manageable number of if-then rules with a small number of premises while maintaining high inferential capacity for all datasets. In turn, argument-based models achieved comparable results to those derived directly from if-then rules, leveraging a concise set of rules and excelling in explainability. In summary, this paper introduces a novel approach for efficiently and automatically generating arguments and their interactions from data, addressing both scientific and applied challenges in advancing the application and deployment of argumentation systems in XAI.

Explainable Artificial Intelligence (XAI) is a research area that clarifies AI decision-making processes to build user trust and promote responsible AI. Hence, a key scientific challenge in XAI is the development of methods that generate transparent and interpretable explanations while maintaining scalability and effectiveness in complex scenarios. Rule-based methods in XAI generate rules that can potentially explain AI inferences, yet they can also become convoluted in large scenarios, hindering their readability and scalability. Moreover, they often lack contrastive explanations, leaving users uncertain why specific predictions are preferred. To address this scientific problem, we explore the integration of computational argumentation—a sub-field of AI that models reasoning processes through defeasibility—into rule-based XAI systems. Computational argumentation enables arguments modelled from rules to be retracted based on new evidence. This makes it a promising approach to enhancing rule-based methods for creating more explainable AI systems. Nonetheless, research on their integration remains limited despite the appealing properties of rule-based systems and computational argumentation. Therefore, this study also addresses the applied challenge of implementing such an integration within practical AI tools. The study employs the Logic Learning Machine (LLM), a specific rule-extraction technique, and presents a modular design that integrates input rules into a structured argumentation framework using state-of-the-art computational argumentation methods. Experiments conducted on binary classification problems using various datasets from the UCI Machine Learning Repository demonstrate the effectiveness of this integration. The LLM technique excelled in producing a manageable number of if-then rules with a small number of premises while maintaining high inferential capacity for all datasets. In turn, argument-based models achieved comparable results to those derived directly from if-then rules, leveraging a concise set of rules and excelling in explainability. In summary, this paper introduces a novel approach for efficiently and automatically generating arguments and their interactions from data, addressing both scientific and applied challenges in advancing the application and deployment of argumentation systems in XAI. [ABSTRACT FROM AUTHOR]

It is often necessary and reasonable to justify preferences before reasoning from them. Moreover, justifying a preference ordering is reduced to justifying the criterion that produces the ordering. This paper builds on the well-known ASPIC+ formalism to develop a model that integrates justifying qualitative preferences with reasoning from the justified preferences. We first introduce a notion of preference criterion in order to model the way in which preferences are justified by an argumentation framework. We also adapt the notion of argumentation theory to build a sequence of argumentation frameworks, in which an argumentation framework justifies preferences that are to underlie the next framework. That is, in our formalism, preferences become not only an input of an argumentation framework, but also an output of it. This kind of input-output process can be applied in the further steps of argumentation. We also explore some interesting properties of our formalism. [ABSTRACT FROM AUTHOR]

Cancellability is an essential feature of implicatures. However, its reliability has been challenged by several cases and examples in which conversational implicatures seem to be hard or even impossible to cancel. Should it then be concluded that not all implicatures are cancellable, and therefore Grice's cancellability test should be weakened or abandoned? The present paper addresses this problem by drawing a distinction between theoretical and practical cancellability, where the latter concept captures the (un)reasonableness of explicit or contextual cancellation. By analyzing three legal cases of misleading advertising, I show how the indefensibility of an interpretation can account for the practical uncancellability of an implicature. To explain and account for the reasonableness of an interpretation, the presumptive inferences involved must be reconstructed and evaluated considering different dimensions, such as the conversational context and the associated presumptions. • Assessing the cancellability test. • Distinguishing between different dimensions of cancellability • Analyzing legal cases of misleading advertising. • Proposing an analytical method for justifying an interpretation. • Providing criteria for determining practical uncancellability [ABSTRACT FROM AUTHOR]

In this paper, we introduce the subject of the special issue Trends in Argumentation Logic. Here we mainly describe two approaches to argumentation logic with explicating monotonic and nonmonotonic, or defeasible, reasoning and explain the role of artificial intelligence in applying argumentation logic. Then we give a short overview of the papers contributed to the special issue. [ABSTRACT FROM AUTHOR]

Different formalisms for defeasible reasoning have been used to represent knowledge and reason in the legal field. In this work, we provide an overview of the following logic-based approaches to defeasible reasoning: defeasible logic, Answer Set Programming, ABA+, ASPIC+, and DeLP. We compare features of these approaches under three perspectives: the logical model (knowledge representation), the method (computational mechanisms), and the technology (available software resources). On top of that, two real examples in the legal domain are designed and implemented in ASPIC+ to showcase the benefit of an argumentation approach in real-world domains. The CrossJustice and Interlex projects are taken as a testbed, and experiments are conducted with the Arg2P technology.

Introduction: The paper investigates the impact of the use of metaphors in reasoning tasks concerning vaccination, especially for defeasible reasoning cases. We assumed that both metaphor and defeasible reasoning can be relevant to let people understand vaccination as an important collective health phenomenon, by anticipating possible defeating conditions. Methods: We hypothesized that extended metaphor could improve both the argumentative and the communicative effects of the message. We designed an empirical study to test our main hypotheses: participants (N = 196, 78% females; Meanage = 27.97 years, SDage = 10.40) were presented with a text about vaccination, described in either literal or metaphorical terms, based on uncertain vs. safe reasoning scenarios. Results: The results of the study confirmed that defeasible reasoning is relevant for the communicative impact of a text and that an extended metaphor enhances the overall communicative effects of the message, in terms of understandability, persuasion, perceived safety, and feeling of control over the health situation, collective trust in expertise and uptake of experts' advice. However, the results show that this effect is significantly nuanced by the type of defeasible reasoning, especially in the case of participants' trust in expertise and commitment to experts' advice. Conclusion: Both communicative and defeasible reasoning competences are needed to enhance trust in immunization, with possible different outcomes at an individual and collective level. [ABSTRACT FROM AUTHOR]

IntroductionThe paper investigates the impact of the use of metaphors in reasoning tasks concerning vaccination, especially for defeasible reasoning cases. We assumed that both metaphor and defeasible reasoning can be relevant to let people understand vaccination as an important collective health phenomenon, by anticipating possible defeating conditions.MethodsWe hypothesized that extended metaphor could improve both the argumentative and the communicative effects of the message. We designed an empirical study to test our main hypotheses: participants (N = 196, 78% females; Meanage = 27.97 years, SDage = 10.40) were presented with a text about vaccination, described in either literal or metaphorical terms, based on uncertain vs. safe reasoning scenarios.ResultsThe results of the study confirmed that defeasible reasoning is relevant for the communicative impact of a text and that an extended metaphor enhances the overall communicative effects of the message, in terms of understandability, persuasion, perceived safety, and feeling of control over the health situation, collective trust in expertise and uptake of experts' advice. However, the results show that this effect is significantly nuanced by the type of defeasible reasoning, especially in the case of participants' trust in expertise and commitment to experts' advice.ConclusionBoth communicative and defeasible reasoning competences are needed to enhance trust in immunization, with possible different outcomes at an individual and collective level.

The hypothesis-driven methodology is a cognitive activity used in expertise processes to solve problems with limited knowledge and understanding. Although some organizations have standardized this approach to guide humans in carrying out expertise in enterprises, it lacks appropriate tools to assist experts in carrying out this cognitive activity, tracking understanding, or capturing the reasoning steps and the knowledge produced during the process. To acquire, share and reuse experts' knowledge applied during expertise processes while assisting humans in bringing understanding to complex problems, this study introduces a human–machine collaborative framework that formalizes experts' knowledge from the hypothesis-driven methodology described in the France standard NF X50-110 of "Quality of expertise activity". This framework utilizes Hypothesis Theory extended with qualitative doubt and a systematic reasoning process to generate a hypothesis exploratory graph (HEG). The proposed approach makes it easier to carry out expertise processes through a human–machine collaboration, offers a means to share and reuse knowledge from expertise, and provides expertise processes evaluation mechanisms. Furthermore, an experiment conducted on a use-case of expertise process verifies the feasibility and effectiveness of the approach. [ABSTRACT FROM AUTHOR]

Introduction. This article focuses on the specifics of the arguments from ignorance and arguments from silence. The relevance of the work is due to the growing interest of the scientific community in modeling of defeasible (presumptive) reasoning and verification of their validity.Methodology and sources. Methodologically, the work is based on the dialectical approach of D. Walton, who proposed a corpus of argumentation schemes for presumptive arguments and a methodology for testing their validity using critical questions. To test the validity of arguments from ignorance and silence, we propose the concept of epistemic burden, the proper fulfillment of which is a necessary condition for the validity of these arguments.Results and discussion. The result of the research is the development of corpuses of critical questions for two schemes of argumentation: the argument for ignorance and the argument from silence, which include questions to check the fulfillment of the epistemic burden by the proponent or information source. The methodology for verifying arguments using the proposed critical questions is analyzed using examples.Conclusion. The difference between the two types of reasoning is that in the argument from ignorance, the epistemic burden is placed on the discussion participants (panelists), and in the argument from silence it is placed on the source of the information.

Different formalisms for defeasible reasoning have been used to represent knowledge and reason in the legal field. In this work, we provide an overview of the following logic-based approaches to defeasible reasoning: defeasible logic, Answer Set Programming, ABA+, ASPIC+, and DeLP. We compare features of these approaches under three perspectives: the logical model (knowledge representation), the method (computational mechanisms), and the technology (available software resources). On top of that, two real examples in the legal domain are designed and implemented in ASPIC+ to showcase the benefit of an argumentation approach in real-world domains. The CrossJustice and Interlex projects are taken as a testbed, and experiments are conducted with the Arg2P technology. [ABSTRACT FROM AUTHOR]

Human conditional reasoning is defeasible: people withdrawlogically valid conclusions if they are aware of situations (i.e.,exceptions) that prevent the consequent of the rule to happenalthough the antecedent is given. In this paper we investigatedefeasible reasoning with quantified rules. In two experimentswe rephrased conditionals from the literature (Experiment 1)and rules from penal code (Experiment 2) as either universalor existential rules and embedded them into Modus Ponensand Modus Tollens inference problems. We show that defeasiblereasoning also exists for quantified rules. However, thekind of quantifier (universal vs. existential) did not affect inferences.This last finding conflicts with theories highlightingthe importance of logic in human reasoning.

We extend universal moral grammar theory (UMGT) with nonmonotoniclogic. Our experiment shows that such revision isnecessary as it allows to account for the effects of alleviationsand aggravations in moral reasoning. Our new theory updatesUMGT from classical to non-monotonic logic, which reflectsthe incompleteness of information and uncertainty in actualhuman reasoning. In addition, it provides an explanation ofthe paradoxical findings in the moral dilemma of the Trolleyproblem and the Knobe effect

The study of defeasible reasoning unites epistemologists with those working in AI, in part, because both are interested in epistemic rationality. While it is traditionally thought to govern the formation and (with)holding of beliefs, epistemic rationality may also apply to the interrogative attitudes associated with our core epistemic practice of inquiry, such as wondering, investigating, and curiosity. Since generally intelligent systems should be capable of rational inquiry, AI researchers have a natural interest in the norms that govern interrogative attitudes. Following its recent coinage, we use the term "zetetic" to refer to the properties and norms associated with the capacity to inquire. In this paper, we argue that zetetic norms can be modeled via defeasible inferences to and from questions -- a.k.a erotetic inferences -- in a manner similar to the way norms of epistemic rationality are represented by defeasible inference rules. We offer a sequent calculus that accommodates the unique features of "erotetic defeat" and that exhibits the computational properties needed to inform the design of zetetic agents. The calculus presented here is an improved version of the one presented in (Millson, 2019), extended to cover a new class of defeasible erotetic inferences. [ABSTRACT FROM AUTHOR]

Deontic logic is a logic often used to formalise scenarios in the legal domain. Within the legal domain there are many exceptions and conflicting obligations. This motivates the enrichment of deontic logic with not only the notion of defeasibility, which allows for reasoning about exceptions, but a stronger notion of typicality that is based on defeasibility. KLM-style defeasible reasoning is a logic system that employs defeasibility while Propositional Typicality Logic (PTL) is a logic that does the same for the notion of typicality. Deontic paradoxes are often used to examine logic systems as the paradoxes provide undesirable results even if the scenarios seem intuitive. Forrester's paradox is one of the most famous of these paradoxes. This paper shows that KLM-style defeasible reasoning and PTL can be used to represent and reason with Forrester's paradox in such a way as to block undesirable conclusions without completely sacrificing desirable deontic properties. [ABSTRACT FROM AUTHOR]

Datalog is a declarative logic programming language that uses classical logical reasoning as its basic form of reasoning. Defeasible reasoning is a form of non-classical reasoning that is able to deal with exceptions to general assertions in a formal manner. The KLM approach to defeasible reasoning is an axiomatic approach based on the concept of plausible inference. Since Datalog uses classical reasoning, it is currently not able to handle defeasible implications and exceptions. We aim to extend the expressivity of Datalog by incorporating KLM-style defeasible reasoning into classical Datalog. We present a systematic approach for extending the KLM properties and a well-known form of defeasible entailment: Rational Closure. We conclude by exploring Datalog extensions of less conservative forms of defeasible entailment: Relevant and Lexicographic Closure. We provide algorithmic definitions for these forms of defeasible entailment and prove that the definitions are LM-rational. [ABSTRACT FROM AUTHOR]

This paper deals with various ways of criticizing arguments. In particular, I focus on John Pollock's theory of defeasible reasoning and show that his conception of rebutting and undercutting defeaters omits some important ways of attacking arguments. I discuss not only the tenets of Pollock's theory but I compare it also to some traditional ways of criticizing arguments occurring in philosophical logic, informal logic and argumentation theory. The paper provides some interesting kinds of counterexamples to deductive arguments which play a similar role to those of the undercutting and the rebutting defeaters occurring in case of non-deductive arguments. Finally, I point to a dilemma emerging from the analysis of Pollock's conception: Either we acknowledge that Pollock's theory is applicable only to a limited scope of defeasible arguments or we admit that we have to broaden the definition of defeasible arguments in such a way that it covers also deductive arguments. However, the latter option can lead to a conclusion that there are no nondefeasible arguments. [ABSTRACT FROM AUTHOR]

Highlights • Collaborative discussion and knowledge acquisition strategy for distributed agents. • Agents end the discussion with the same assessment about the shared information. • Agents can freely and dynamically join or leave the discussion. • Structured argumentation is used for knowledge representation and reasoning. • The strategy is formally characterized and empirically analysed. Abstract We present a strategy for collaborative reasoning and automatic knowledge sharing in which multiple agents discuss about some piece of information until they reach an agreement. A structured argumentation system (DeLP) is used as the underlying knowledge representation and reasoning formalism, allowing agents to contribute to the discussion by posing or rejecting arguments, or objecting to atomic information. In this proposal, in order to be able to represent more general scenarios than other approaches in the literature, agents are allowed to believe in strict information and are allowed to freely join and leave the discussion at any moment. The proposed strategy automatically deals with the conflicts that may arise when agents pose contradictory strict information, which have to be solved to maintain the soundness of the discussion. The operational semantics of the strategy is defined using a transition system to formally characterise the agents' behaviour and to prove its properties. An empirical comparison between the performance of different criteria that the agents can follow during the discussion is presented. [ABSTRACT FROM AUTHOR]

In this paper, we extend a temporal defeasible logic with a modal operator Committed to formalize commitments that agents undertake as a consequence of communicative actions (speech acts) during dialogues. We represent commitments as modal sentences. The defeasible dual of the modal operator Committed is a modal operator called Exempted. The logical setting makes the social-commitment based semantics of speech acts verifiable and practical; it is possible to detect if, and when, a commitment is violated and/or complied with. One of the main advantages of the proposed system is that it allows for capturing the nonmonotonic behavior of the commitments induced by the relevant speech acts. [ABSTRACT FROM AUTHOR]

This contribution aims at providing an argumentative method to account for epistemic modality and evidentiality. I claim that these two linguistic categories can work as semantic components of defeasible argumentative schemes based on classification processes. This kind of approximate reasoning is, in fact, frequently indicated by appearance verbs which signal that the inferred standpoint is conceived by the speaker as uncertain (epistemic value) due to the deceiving nature of perceptual data (evidential value). Drawing from an analysis at the semantic-argumentative interface, the way in which prototype theory sheds light on the processes of meaning construction underlying defeasible arguments from definition is also shown.

This article considers defeasible beliefs in dynamic settings. In particular, we examine the belief projection problem: what is believed after performing an action and/or receiving new information? The approach is based on an epistemic variant of Reiter's situation calculus, where actions not only have physical effects but may also provide new information to the agent. The preferential belief structure is initially determined using conditional statements. New information is then incorporated using two popular belief revision schemes, namely natural and lexicographic revision. The projection problem is solved twofold in this formalism: by goal regression and by knowledge base progression. [ABSTRACT FROM AUTHOR]

The principle of reinstatement governing most argument systems states that an argument is reinstated when all its defeaters are in turn ultimately defeated. Nevertheless, some criticisms to this principle have been offered in the literature. We found that problems arise when arguments in a chain of attacks are related by specificity: when non-maximally specific arguments are reinstated, fallacious justifications are originated. Particularly, we show how the problem affects DeLP, a system that combines a specificity-based defeat criterion with a reinstatement-based warrant process. Following old intuitions by philosopher Carl Hempel we rethink the concept's role within defeasible argumentation. Two kinds of specificity defeaters are identified: proper defeaters and cautious defeaters. While proper defeaters are well-known, cautious defeaters are formally introduced here. A system combining cautious and proper defeaters is defined as an extension of DeLP, and dialectic warrant games are proposed for filtering out non-maximally specific arguments. [ABSTRACT FROM AUTHOR]

Copyright of Praxis Filosófica is the property of Universidad del Valle and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

DL N is a recent nonmonotonic description logic, designed for satisfying independently proposed knowledge engineering requirements, and for removing some recurrent drawbacks of traditional nonmonotonic semantics. In this paper we study the logical properties of DL N and illustrate some of the relationships between the KLM postulates and the characteristic features of DL N , including its novel way of dealing with unresolved conflicts between defeasible axioms. Moreover, we fix a problem affecting the original semantics of DL N and accordingly adapt the reduction from DL N inferences to classical inferences. Along the paper, we use various versions of the KLM postulates to deepen the comparison with related work, and illustrate the different tradeoffs between opposite requirements adopted by each approach. [ABSTRACT FROM AUTHOR]

Conditionals are useful for modelling many forms of everyday human reasoning but are not always sufficiently expressive to represent the information we want to reason about. In this paper, we make a case for a form of situated conditional. By 'situated', we mean that there is a context, based on an agent's beliefs and expectations, that works as background information in evaluating a conditional, and we allow such a context to vary. These conditionals are able to distinguish, for example, between expectations and counterfactuals. Formally, they are shown to generalise the conditional setting in the style of Kraus, Lehmann, and Magidor. We show that situated conditionals can be described in terms of a set of rationality postulates. We then propose an intuitive semantics for these conditionals and present a representation result which shows that our semantic construction corresponds exactly to the description in terms of postulates. With the semantics in place, we define a form of entailment for situated conditional knowledge bases, which we refer to as minimal closure. Finally, we proceed to show that it is possible to reduce the computation of minimal closure to a series of propositional entailment and satisfiability checks. While this is also the case for rational closure, it is somewhat surprising that the result carries over to minimal closure. [ABSTRACT FROM AUTHOR]