Dialectical Explanations in Defeasible Argumentation

An argument is reinstated when all its defeaters are in turn ultimately defeated. This is a kind of principle governing most argument systems in AI. Nevertheless, some criticisms to this principle have been offered in the literature. Assuming that reinstatement is prima facie acceptable, we analyze some counterexamples in order to identify common causes. As a result, we found that the problem arises when arguments in a chain of attacks are related by specificity. We argue that the reason is that non-maximally specific arguments can be reinstated originating fallacious justifications. Following old intuitions by Carl Hempel about inductive explanations, we propose a requirement of maximal specificity on defeasible arguments and introduce “undermining defeaters” which, in essence, facilitate the rejection of those arguments which do not satisfy the requirement. This ideas are formally defined using the DeLP system for defeasible logic programming.KeywordsArgument StructureWinning StrategyArgumentation FrameworkTotal EvidenceNonmonotonic ReasoningThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Inspired by legal reasoning, this paper presents a semantics and proof theory of a system for defeasible argumentation. Arguments are expressed in a logic-programming language with both weak and strong negation, conflicts between arguments are decided with the help of priorities on the rules. An important feature of the system is that these priorities are not fixed, but are themselves defeasibly derived as conclusions within the system. Thus debates on the choice between conflicting arguments can also be modelled. The semantics of the system is given with a fixpoint definition, while its proof theory is stated in dialectical style, where a proof takes the form of a dialogue between a proponent and an opponent of an argument: an argument is shown to be justified if the proponent can make the opponent run out of moves in whatever way the opponent attacks.

Argumentation systems can be employed for detecting spatiotemporal patterns. While the idea of argumentation consists in defending specific positions, complex patterns are influenced by several factors that can be regarded as arguments against or in favor of the realisation of those patterns. The idea is to determine consistent positions of arguments which speak for specific patterns. This becomes possible by means of algorithms which have been defined for argumentation systems. The introduced method of conceptual argumentation is new in comparison to classical, i.e. value-based, argumentation systems. It has the advantage to be more flexible by enabling the definition of conceptual arguments influencing relevant patterns. There are two main results: first, conceptual argumentation frameworks do scale significantly better; secondly, investigating our approach by examining soccer games, we show that specific patterns, such as passes, can be detected with different retrieval performances depending on the chosen spatial granularity level.KeywordsDescription LogicSemantic DescriptionArgumentation FrameworkDynamic SceneDefault TheoryThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Two important notions within the field of classical argumentation are undercutting defeaters and backings. The former represent an attack to an inference step, and the latter intend to provide defense against this type of attack. Defeasible Logic Programming (DELP) is a concrete argumentation system that allows to identify arguments whose conclusions or intermediate conclusions are in contradiction, capturing the notion of rebutting defeater. Nevertheless, in DELP is not possible to represent neither undercutting defeaters nor backings. The aim of this work is to extend the formalism of DELP to allow attack and support for defeasible rules. Thus, it will be possible to build arguments for representing undercutting defeaters and backings.

We present an object-oriented method to represent knowledge about a set of digital systems. Based on this representation, concrete systems can easily be specified, their implementations in various models such as VHDL or Petrinets can be derived automatically from a specification by a knowledge based configuration system. In this paper, we focus on the set-theoretic foundations of our model.KeywordsDomain ModelFunction LevelConceptual HierarchyInstruction CacheConcrete SystemThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

This article investigates to what extent protocols for dynamicdisputes, i.e., disputes in which the information base can vary at differentstages, can be justified in terms of logics for defeasible argumentation. Firsta general framework is formulated for dialectical proof theories for suchlogics. Then this framework is adapted to serve as a framework for protocols fordynamic disputes, after which soundness and fairness properties are formulated for such protocols relative to dialectical proof theories. It then turns out that certaintypes of protocols that are perfectly fine with a static information base, arenot sound or fair in a dynamic setting. Finally, a natural dynamic protocolis defined for which soundness and fairness can be established.

The traditional approach to automatic verification of value-passing processes is to translate them into pure processes, by instantiating input variables over all possible values, then submit the resulting pure processes to verification tools. The main disadvantage of this approach is that problems with infinite value domain can not be dealt with, even if these values are never used by the program, as in the case of communication protocols. In this paper we propose an algorithm which works directly on value-passing processes, represented as symbolic transition graphs with assignment. The key idea is to instantiate input variables “on-the-fly”, while checking bisimulation, making it possible not to store the entire concrete transition system either before or during the verification process. As a consequence problems with large, even infinite, state spaces can be handled. The algorithm has been implemented and tested on some typical verification problems.KeywordsOperational SemanticState PairTransition GraphInput ActionLabel Transition SystemThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Every product system E is associated with a natural Hilbert space L2(E). There is also a natural way of representing E as a concrete product system ε acting on L2(E), and ε is isomorphic to E. In turn, ε is associated with a semigroup of endomorphisms α= {α t : t≥0} of B(L2(E)) as in Proposition 2.4.9. However, this semigroup is not an E0-semigroup because the projections α t (1) decrease to 0 as t → ∞. This chapter focuses on the problem of constructing an E0-semigroup whose product system is isomorphic to E.KeywordsHilbert SpaceProduct SystemOrthonormal BasisRegular RepresentationApproximate UnitThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Abstract argumentation systems are formalisms for defeasible reasoning where some components remain unspecified, the structure of arguments being the main abstraction. In the dialectical process carried out to identify accepted arguments in the system some controversial situations may appear. These relate to the reintroduction of arguments into the process which cause the onset of circularity. This must be avoided in order to prevent an infinite analysis. Some systems apply the sole restriction of not allowing the introduction of previously considered arguments in an argumentation line. However, repeating an argument is not the only possible cause for the risk mentioned. A more specific restriction needs to be applied considering the existence of subarguments. In this work, we introduce an extended argumentation framework where two kinds of defeat relation are present, and a definition for progressive defeat path.KeywordsAbstract ArgumentationArgumentation FrameworkAbstract FrameworkPreference CriterionDefeasible ReasoningThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Improving remote sensing crop classification by argumentation-based conflict resolution in ensemble learning

Ostrovsky et al. [1] gave the first definition of non-malleable witness-indistinguishable argument systems. A surprising result given by them showed this notion was incomparable with the notion of non-malleable zero-knowledge. However, they only discussed their relations in the interactive setting. In this paper, we make an observation on relation between the two notions in the non-interactive setting. We show the two notions are still incomparable: that is, there are non-malleable non-interactive zero-knowledge proof systems that are not non-malleable non-interactive witness-indistinguishable, and vice versa.KeywordsProof SystemCommitment SchemeCommon Reference StringTrapdoor PermutationPlain ModelThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

The combinatorics of proof-search in classical propositional logic lies at the heart of most efficient proof procedures because the logic admits least-commitment search. The key to extending such methods to quantifiers and non-classical connectives is the problem of recovering this least-commitment principle in the context of the non-classical/non-propositional logic; i.e., characterizing when a least-commitment (classical) search yields sufficient evidence for provability in the (non-classical) logic.In this paper, we present such a characterization for the (⊃, ∧)-fragment of intuitionistic logic using the λμ-calculus: a system of realizers for classical free deduction (cf. natural deduction) due to Parigot.We show how this characterization can be used to define a notion of uniform proof, and a corresponding proof procedure, which extends that of Miller et al. to multiple-conclusioned sequent systems. The procedure is sound and complete for the fragment of intuitionistic logic considered and enjoys the combinatorial advantages of search in classical logic.KeywordsClassical LogicIntuitionistic LogicNatural DeductionSequent CalculusProof ProcedureThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

We define an automatic proof procedure for finitely many-valued logics given by truth tables. The proof procedure is based on the inverse method. To define this procedure, we introduce so-called introduction-based sequent calculi. By studying proof-theoretic properties of these calculi we derive efficient validity- and satisfiability-checking procedures based on the inverse method. We also show how to translate the validity problem for a formula to unsatisfiability checking of a set of propositional clauses.KeywordsInference RuleTheorem ProveInverse MethodAutomate ReasoningPropositional VariableThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

We consider the product-ranking challenge that online retailers face when their customers typically behave as “window shoppers.” They form an impression of the assortment after browsing products ranked in the initial positions and then decide whether to continue browsing. We design online learning algorithms for product ranking that maximize the number of customers who engage with the site. Customers’ product preferences and attention spans are correlated and unknown to the retailer; furthermore, the retailer cannot exploit similarities across products, owing to the fact that the products are not necessarily characterized by a set of attributes. We develop a class of online learning-then-earning algorithms that prescribe a ranking to offer each customer, learning from preceding customers’ clickstream data to offer better rankings to subsequent customers. Our algorithms balance product popularity with diversity, the notion of appealing to a large variety of heterogeneous customers. We prove that our learning algorithms converge to a ranking that matches the best-known approximation factors for the offline, complete information setting. Finally, we partner with Wayfair — a multibillion-dollar home goods online retailer — to estimate the impact of our algorithms in practice via simulations using actual clickstream data, and we find that our algorithms yield a significant increase (5–30%) in the number of customers that engage with the site. This paper was accepted by J. George Shanthikumar, Management Science Special Section on Data-Driven Prescriptive Analytics.

All the algorithms described in the chapters 2, 3, 4 and 6, viz., FC-MLP, IC-MLP, CRBF, FC-RBF, Mc-FCRBF, FCRN and CC-ELM are batch learning algorithms. These algorithms require the complete training data set and the network structure has to be fixed a priori.KeywordsExtreme Learning MachineHide NeuronRadial Basis Function Neural NetworkNetwork ParameterTraining SequenceThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.