Probabilistic abstract argumentation frameworks, a possible world view

Mantadelis, Theofrastos; Bistarelli, Stefano

doi:10.1016/j.ijar.2019.12.006

Cited by 15 publications

(10 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The admissible arguments are defined by the classical extension-based semantics (Dung 1995). Mantadelis and Bistarelli (2020) presents an encoding of a constellations approach in ProbLog. The epistemic approach interprets the probabilities as a direct measure of an agent's belief in the arguments.…”

Section: Related Workmentioning

confidence: 99%

“…This question has been only partially investigated so far: of the two main interpretations of probabilities in argument graphs (Hunter 2013), namely the constellations approach and the epistemic approach, only the former has been studied in the context of PLP (Mantadelis and Bistarelli 2020). In fact, probabilistic argumentation systems propose different combinations of argumentation frameworks, probability interpretations and reasoning systems, tailored to manipulating probabilities in argumentation.…”

Section: Introductionmentioning

confidence: 99%

“…In fact, probabilistic argumentation systems propose different combinations of argumentation frameworks, probability interpretations and reasoning systems, tailored to manipulating probabilities in argumentation. As we summarize in Table 1, some do not follow the epistemic approach (Li, Oren, and Norman 2011;Kido and Okamoto 2017;Mantadelis and Bistarelli 2020), others do not consider PLP or a Bayesian view (i.e. causal influence between random variables defining a single probability distribution) of probabilities (Hunter, Polberg, and Thimm 2020).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

SMProbLog: Stable Model Semantics in ProbLog and its Applications in Argumentation

Totis¹,

Kimmig²,

Raedt³

2021

Preprint

View full text Add to dashboard Cite

We introduce SMProbLog, a generalization of the probabilistic logic programming language ProbLog. A ProbLog program defines a distribution over logic programs by specifying for each clause the probability that it belongs to a randomly sampled program, and these probabilities are mutually independent. The semantics of ProbLog is given by the success probability of a query, which corresponds to the probability that the query succeeds in a randomly sampled program. It is well-defined when each random sample uniquely determines the truth values of all logical atoms. Argumentation problems, however, represent an interesting practical application where this is not always the case. SMProbLog generalizes the semantics of ProbLog to the setting where multiple truth assignments are possible for a randomly sampled program, and implements the corresponding algorithms for both inference and learning tasks. We then show how this novel framework can be used to reason about probabilistic argumentation problems. Therefore, the key contribution of this paper are: a more general semantics for ProbLog programs, its implementation into a probabilistic programming framework for both inference and parameter learning, and a novel approach to probabilistic argumentation problems based on such framework.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

SMProbLog: Stable Model Semantics in ProbLog and its Applications in Argumentation

Totis¹,

Kimmig²,

Raedt³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Different authors provide AAF, which allow the consideration of argument value (Bench-Capon, 2003) or weight of an argument (Dunne et al, 2011). In parallel, works have been done to extend the Dung model by assigning probabilities to argument attacks or extensions (Grossi and Hoek, 2013;Hunter, 2013Hunter, , 2014Li et al, 2011;Rienstra, 2012;Mantadelis and Bistarelli, 2020). The two main approaches for adding probabilistic information to the AAF are the constellation (Hunter, 2014;Li et al, 2011) and the epistemic approaches (Polberg et al, 2017;Thimm, 2012).…”

Section: Imprecise Abstract Argumentationmentioning

confidence: 99%

Imprecise abstract argumentation as a support for forensic engineering

Taillandier

Baudrit

Carvajal

et al. 2021

ECAM

View full text Add to dashboard Cite

PurposeCivil engineering structures are regularly confronted with failures that can lead to catastrophic consequences. It is important, after a failure, to be able to identify the origin and the sequence of factors that led to it. This failure analysis by experts, called forensic engineering investigation, generally leads to the drafting of an expert report. These reports do not inform on the processes that guided the experts to a conclusion and the uncertainties involved. This paper aims to propose a new methodological approach to formalize the opinions of experts in forensic engineering.Design/methodology/approachThe research consists in combining abstract argumentation with the theory of imprecise probabilities to take into account epistemic and stochastic uncertainties to support forensic engineering investigation.FindingsA model and a tool to support forensic analysis are presented. An application on the collapse of the Brumadinho dam highlights the interest of the chosen approach.Originality/valueThis work is the first use of the abstract argument framework in civil engineering, and so in forensic engineering. Furthermore, it provides an innovative model based on imprecise probability for AAF.

show abstract

“…Besides neuro-symbolic argumentation systems, various systems have been investigated to determine the probability of some argument statuses given the statuses of some other arguments or premises, with respect to sundry probabilistic settings (often with no learning abilities), see e.g. (Riveret et al 2007;Fazzinga, Flesca, and Parisi 2016;Potyka 2019;Mantadelis and Bistarelli 2020). Usually such works have strong assumptions on probabilistic dependencies (typically arguments are assumed to be probabilistically independent) while we do not rely on such assumptions thanks to the probabilistic graphical model of RBMs.…”

Section: Related Workmentioning

confidence: 99%

Neuro-Symbolic Probabilistic Argumentation Machines

Riveret

Tran

Garcez

2020

Proceedings of the Seventeenth International Conference on Principles of Knowledge Representation and Reasoning

View full text Add to dashboard Cite

Neural-symbolic systems combine the strengths of neural networks and symbolic formalisms. In this paper, we introduce a neural-symbolic system which combines restricted Boltzmann machines and probabilistic semi-abstract argumentation. We propose to train networks on argument labellings explaining the data, so that any sampled data outcome is associated with an argument labelling. Argument labellings are integrated as constraints within restricted Boltzmann machines, so that the neural networks are used to learn probabilistic dependencies amongst argument labels. Given a dataset and an argumentation graph as prior knowledge, for every example/case K in the dataset, we use a so-called K-maxconsistent labelling of the graph, and an explanation of case K refers to a K-maxconsistent labelling of the given argumentation graph. The abilities of the proposed system to predict correct labellings were evaluated and compared with standard machine learning techniques. Experiments revealed that such argumentation Boltzmann machines can outperform other classification models, especially in noisy settings.

show abstract

Probabilistic abstract argumentation frameworks, a possible world view

Cited by 15 publications

References 31 publications

SMProbLog: Stable Model Semantics in ProbLog and its Applications in Argumentation

SMProbLog: Stable Model Semantics in ProbLog and its Applications in Argumentation

Imprecise abstract argumentation as a support for forensic engineering

Neuro-Symbolic Probabilistic Argumentation Machines

Contact Info

Product

Resources

About