Summary Report for the Third International Competition On Computational Models of Argumentation

The Third International Competition on Computational Models of Argumentation (ICCMA’19) focused on reasoning tasks in abstract argumentation frameworks. Submitted solvers were tested on a selected collection of benchmark instances, including artificially generated argumentation frameworks and some frameworks formalizing real-world problems. This competition introduced two main novelties over the two previous editions: the first one is the use of the Docker platform for packaging the participating solvers into virtual “light” containers; the second novelty consists of a new track for dynamic frameworks.

Conflict Detection Predicts the Temporal Stability of Intuitive and Deliberate Reasoning

Although the susceptibility to reasoning biases is often assumed to be a stable trait, the temporal stability of people’s performance on popular heuristics-and-biases tasks has been rarely directly tested. The present study addressed this issue and examined a potential determinant for answer change. Participants solved the same set of “bias” tasks twice in two test sessions, two weeks apart. We used the two-response paradigm to test the stability of both initial (intuitive) and final (deliberate) responses. We hypothesized that participants who showed higher conflict detection in their initial intuitive responses at session 1 (as indexed by a relative confidence decrease compared to control problems), would be less stable in their responses between session 1 and 2. Results showed that performance on the reasoning tasks was highly, but not entirely, stable two weeks later. Notably, conflict detection in session 1 was significantly more pronounced in those cases that participants did change their answer between sessions. We discuss practical and theoretical implications.

Harnessing Incremental Answer Set Solving for Reasoning in Assumption-Based Argumentation

Assumption-based argumentation (ABA) is a central structured argumentation formalism. As shown recently, answer set programming (ASP) enables efficiently solving NP-hard reasoning tasks of ABA in practice, in particular in the commonly studied logic programming fragment of ABA. In this work, we harness recent advances in incremental ASP solving for developing effective algorithms for reasoning tasks in the logic programming fragment of ABA that are presumably hard for the second level of the polynomial hierarchy, including skeptical reasoning under preferred semantics as well as preferential reasoning. In particular, we develop non-trivial counterexample-guided abstraction refinement procedures based on incremental ASP solving for these tasks. We also show empirically that the procedures are significantly more effective than previously proposed algorithms for the tasks.

Existential Abstraction on Argumentation Frameworks via Clustering

Argumentation in Artificial Intelligence (AI) builds on formal approaches to reasoning argumentatively. Common to many such approaches is to use argumentation frameworks (AFs) as reasoning engines, with AFs being composed of arguments and attacks between arguments, which are instantiated from knowledge bases in a principle-based manner. While representing what can be argued for in an AF provides a conceptually clean way, this process can face challenges arising from generating a large number of arguments, which can act as a barrier to explainability. Inspired by successful approaches to model checking where the state explosion is mitigated by applying existential abstraction, we study an adaption of existential abstraction in form of clustering arguments in an AF to address an associated "argument explosion". In this paper, we provide a foundational investigation of this form of existential abstraction by defining semantics of the resulting clustered AFs, which balance two inherent aspects of existential abstractions: abstracting from concrete AFs and not permitting too much spuriousness (i.e., conclusions that hold on the abstraction but not on the original AF). Moreover, we show properties of clustered AFs, including complexity results, discuss use of clusterings for explaining results of reasoning tasks, and employ the recently introduced methodology of abstraction in answer set programming (ASP) for obtaining and reasoning over clustered AFs.

Neighborhood aggregation based graph attention networks for open-world knowledge graph reasoning

Knowledge graph reasoning or completion aims at inferring missing facts based on existing ones in a knowledge graph. In this work, we focus on the problem of open-world knowledge graph reasoning—a task that reasons about entities which are absent from KG at training time (unseen entities). Unfortunately, the performance of most existing reasoning methods on this problem turns out to be unsatisfactory. Recently, some works use graph convolutional networks to obtain the embeddings of unseen entities for prediction tasks. Graph convolutional networks gather information from the entity’s neighborhood, however, they neglect the unequal natures of neighboring nodes. To resolve this issue, we present an attention-based method named as NAKGR, which leverages neighborhood information to generate entities and relations representations. The proposed model is an encoder-decoder architecture. Specifically, the encoder devises an graph attention mechanism to aggregate neighboring nodes’ information with a weighted combination. The decoder employs an energy function to predict the plausibility for each triplets. Benchmark experiments show that NAKGR achieves significant improvements on the open-world reasoning tasks. In addition, our model also performs well on the closed-world reasoning tasks.

084 Does Alignment between the Timing of Sleep and Circadian Rhythm Predict Behavioral Decision Making?

Introduction Cognitive performance and decision making have been shown to suffer under conditions of misalignment between circadian preference and time-of-assessment; however, little is known about how misalignment between the timing of sleep and circadian rhythm impacts decision making. To this end, this study captured naturally occurring degrees of alignment between the timing of sleep and the circadian rhythm (i.e., alignment of sleep-wake timing with circadian phase) to examine if greater misalignment predicts worse behavioral decision making. Methods Over the course of two weeks, 32 participants (18–22 years of age; 61% female; 69% White) continuously wore actigraphs and completed two overnight in-lab visits (Thursday and Sunday) in which both dim light melatonin onset (DLMO) and behavioral decision-making (risk taking, framing, and strategic reasoning tasks) were assessed. Sleep-wake timing was assessed by actigraphic midsleep from the two nights prior to each in-lab visit. Alignment was operationalized as the interval between DLMO and average midsleep. Multilevel modeling was used to predict performance on decision making tasks from circadian alignment during each in-lab visit; nonlinear associations were also examined. Results Misalignment characterized by shorter time between DLMO and midsleep predicted decision-making in a curvilinear fashion (i.e., squared misalignment term predicted performance). Specifically, shorter time between DLMO and midsleep predicted greater risk-taking under conditions of potential loss (B = .10, p = .04), but less risk-taking under conditions of potential reward (B = -.14, p = .04) in a curvilinear fashion. Misalignment did not predict decision-making in the framing and strategic reasoning tasks. Conclusion Findings suggest that naturally occurring degrees of misalignment between the timing of sleep and the circadian rhythm may impact risky decision-making, further extending accumulating evidence that sleep/circadian factors are tied to risk-taking preferences. Future studies will need to replicate findings and experimentally probe whether manipulating alignment influences risky decision making. Support (if any) R21AA023209; R01DA044143

Student Development in Logical Reasoning: Results of an Intervention Guiding Students Through Different Modes of Visual and Formal Representation

Due to growing interest in twenty-first-century skills, and critical thinking as a key element, logical reasoning is gaining increasing attention in mathematics curricula in secondary education. In this study, we report on an analysis of video recordings of student discussions in one class of seven students who were taught with a specially designed course in logical reasoning for non-science students (12th graders). During the course of 10 lessons, students worked on a diversity of logical reasoning tasks: both closed tasks where all premises were provided and everyday reasoning tasks with implicit premises. The structure of the course focused on linking different modes of representation (enactive, iconic, and symbolic), based on the model of concreteness fading (Fyfe et al., 2014). Results show that students easily link concrete situations to certain iconic referents, such as formal (letter) symbols, but need more practice for others, such as Venn and Euler diagrams. We also show that the link with the symbolic mode, i.e. an interpretation with more general and abstract models, is not that strong. This might be due to the limited time spent on further practice. However, in the transition from concrete to symbolic via the iconic mode, students may take a step back to a visual representation, which shows that working on such links is useful for all students. Overall, we conclude that the model of concreteness fading can support education in logical reasoning. One recommendation is to devote sufficient time to establishing links between different types of referents and representations.