10.48448/b893-g812

Relational concept analysis (RCA) is an extension of formal concept analysis dealing with several related formal contexts simultaneously. It can learn description logic theories from data and has been used within various applications. However, RCA returns a single family of concept lattices, though, when the data feature circular dependencies, other solutions may be considered acceptable. The semantics of RCA, provided in an operational way, does not shed light on this issue. This paper aims at defining precisely the semantics of RCA and identifying alternative solutions. We first characterise the acceptable solutions as those families of concept lattices which belong to the space determined by the initial contexts (well-formed), which cannot scale new attributes (saturated), and which refer only to concepts of the family (self-supported). We adopt a functional view on the RCA process by defining the space of well-formed solutions and two functions on that space: one expansive and the other contractive. In this context, the acceptable solutions are the common fixed points of both functions. We show that RCA returns the least element of the set of acceptable solutions. In addition, it is possible to build dually an operation that generates its greatest element. The set of acceptable solutions is a complete sublattice of the interval between these two elements. Its structure, and how the defined functions traverse it, are studied in detail. 83 1 10.1613/jair.1.17882

111 Knowledge is widely distributed within our society. It has to be accessed and combined to be properly exploited. These notes address the problems in dealing with distributed knowledge. They first consider how to represent knowledge and especially classifications or ontologies. Because there may be different such representations, they are aligned in order to create networks of ontologies. These networks enable answering queries based on connected distributed knowledge and data. But their heterogeneity may lead knowledge to be inconsistent or contradictory. Belief revision is then used in order to restore consistency within networks of ontologies. However, other ways to deal with distributed knowledge and its evolution exist. One of these, cultural knowledge evolution, consists of selecting knowledge in an evolutionary way. It is considered experimentally, through multi-agent simulations, and theoretically, through dynamic epistemic logic models. The mentioned topics are not explored in depth. Instead, the tight relationships between these are considered from a semantic standpoint.

Different disciplines have been studying human memory and related issues for thousands of years. However, the definitions of the concepts relating to memory vary depending on the discipline or theory. In order to conciliate these variations and ambiguities, a solution is to formally define the concepts studied through ontologies. This paper presents Mem’Onto, a Memory Ontology which gather concepts related to memory, based on the Tulving’s SPI model. This theory corresponds to a model of memory organisation and brings together various central elements of memory according to Tulving, whether in Memory Systems (e.g. Episodic Memory, Semantic Memory, Procedural Memory), in Mnesic Processes (e.g. Encoding, Storage, Retrieval) or in the level of consciousness of these subsystems during Retrieval (Implicit and Explicit). Mem’Onto is adapted from an existing ontology, CoTOn, a Cognitive Theory Ontology designed from a working memory use case.

Standpoint logic is a recently proposed modal logic framework that is well-suited for multiperspective reasoning and ontology integration. For this reason, combinations of standpoint logic with description logics (DLs), a popular family of logic-based ontology languages, are of special interest. Prior work has shown that it is possible to add standpoints to numerous decidable fragments of first-order logics – including very expressive DLs up to SROIQbs – while preserving their reasoning complexity, so long as standpoint modalities are limited to the axiom level. A more expressive tighter modal integration, where standpoint modalities are also allowed to occur in concept expressions, has so far only been investigated for the much less expressive DL EL+. In this paper, we push this line of research showing that the DL SHIQ allows for a tight modal integration with standpoints without compromising its EXPTIME reasoning complexity. The core insight toward this result is that any satisfiable knowledge base admits a model with only polynomially many worlds, an argument which requires a rather elaborate model-theoretic construction. This allows us to establish a polynomial equisatisfiable translation into plain SHIQ which, beyond showing the theoretical result, enables us to use highly optimised OWL reasoners to provide practical reasoning support for ontology languages extended by standpoint modelling. We complement our findings with the observation that our techniques would fail upon adding the modeling feature of nominals to the underlying DL. 978-1-956792-05-8 10.24963/kr.2024/36 383–393

Standpoint extensions of KR formalisms have been recently introduced to incorporate multi-perspective modelling and reasoning capabilities. In such modal extensions, the integration of conceptual modelling and perspective annotations can be more or less tight, with monodic standpoint extensions striking a good balance as they enable advanced modelling while preserving good reasoning complexities. We consider the extension of C2 — the counting two-variable fragment of first-order logic — by monodic standpoints. At the core of our treatise is a polytime translation of formulae in said formalism into standpoint-free C2, requiring elaborate model-theoretic arguments. By virtue of this translation, the NEXPTIME-complete complexity of checking satisfiability in C2 carries over to our formalism. As our formalism subsumes monodic S5 over C2, our result also significantly advances the state of the art in research on first-order modal logics. As a practical consequence, the very expressive description logics SHOIQBs and SROIQBs which subsume the popular W3C-standardized OWL 1 and OWL 2 ontology languages are shown to allow for monodic standpoint extensions without any increase of standard reasoning complexity. We prove that NEXPTIME-hardness already occurs in much less expressive DLs as long as they feature both nominals and monodic standpoints. We also show that, with inverses, functionality, and nominals present, minimally lifting the monodicity restriction leads to undecidability. 366–375

The paradigm of algebraic constraint-based reasoning, embodied in the notion of a qualitative calculus, is studied within two alternative frameworks. One framework defines a qualitative calculus as "a non-associative relation algebra (NA) with a qualitative representation", the other as "an algebra generated by jointly exhaustive and pairwise disjoint (JEPD) relations". These frameworks provide complementary perspectives: the first is intensional (axiom-based), whereas the second one is extensional (based on semantic structures). However, each definition admits calculi that lie beyond the scope of the other. Thus, a qualitatively representable NA may be incomplete or non-atomic, whereas an algebra generated by JEPD relations may have non-involutive converse and no identity element. The divergence of definitions creates a confusion around the notion of a qualitative calculus and makes the "what" question posed by Ligozat and Renz actual once again. Here we define the relation-type qualitative calculus unifying the intensional and extensional approaches. By introducing the notions of weak identity, inference completeness and Q-homomorphism, we give equivalent definitions of qualitative calculi both intensionally and extensionally. We show that "algebras generated by JEPD relations" and "qualitatively representable NAs" are embedded into the class of relation-type qualitative algebras. 289 103385 10.1016/j.artint.2020.103385

Thèse d'informatique 2022GRALM021 Knowledge Graphs (KGs) are unceasingly used by different organisation to represent real- world entities in the form of a graph. They may use an ontological layer for describing the classes and properties of the represented entities. RDF knowledge graphs are knowledge graphs that convey to the RDF model. RDF knowledge graph interlinking is the task of identifying different IRIs belonging to different RDF knowledge graphs and referring to the same real- world entity. This facilitates data integration and interoperability by combining different entity descriptions present in different knowledge graphs.There exist different methods for addressing the task of interlinking RDF knowledge graph. Link keys are among these methods. They are used for interlinking RDF knowledge graphs described using different ontologies. Link keys specify the properties to be compared to decide whether two entities belonging to different classes and present in different knowledge graphs are the same.Link keys can be expressed as logical axioms, and, thus, it is possible to combine them with ontologies, and ontology alignments to perform logical reasoning. In this thesis, we aim to study the problem of reasoning with link keys. To formally investigate this problem, we model RDF knowledge graphs, ontologies, and ontology alignments using the description logic ALC. We choose the description logic ALC as a base language for reasoning. ALC covers many modeling capabilities used for knowledge representation and allows for a more easy extension to more expressive description logics. We extend ALC with link keys and individual equalities, the resulting description logic is called ALC+LK. We show that link key entailment can be reduced to link key consistency checking without the need of introducing the negation of link keys.Then we design an algorithm for deciding the consistency of ALC+LK ontology. We have proved that the algorithm is sound, complete, and always terminates. This algorithm runs in 2EXPTIME. However, there exist EXPTIME algorithms for reasoning in ALC and the completion rules added for handling link keys and equalities require no more computational power than that of ALC.In the light of the above, we design a sound, complete, worst-case optimal algorithm for reasoning in ALC+LK. This algorithm is inspired by the compressed tableau algorithm, which allows obtaining the EXPTIME optimal complexity result. However, this algorithm has a non- directed behaviour which obstruct its implementation.Last but most importantly, we propose a sound, complete, and worst-case optimal tableau algorithm for reasoning in the description logic ALC with individuals and link keys. This al- gorithm, in contrast to the non-directed one, is directed by the application of completion rules. This avoids the generation of useless structures and facilitates its implementation. We implement this algorithm and provide a number of proof-of-concept experiments that demonstrates the importance of reasoning with link keys for the data interlinking task.

Plusieurs méthodes ont été proposées pour aborder les tâches d’interconnexion de données et d’alignement d’ontologies, qui sont généralement traitées séparément. Dans cet article, nous présentons DICAP, un algorithme qui permet leur collaboration mutuelle. Les expériences réalisées montrent que l’ajout de relations owl:sameAs résultant de l’interconnexion de données permet de découvrir des correspondances ontologiques supplémentaires. De plus, la présence de correspondances ontologiques permet l’extraction de règles de liage supplémentaires et discriminantes. 59–68

Entity Matching (EM) automates the discovery of identity links between entities within different Knowledge Graphs (KGs). Link keys are crucial for EM, serving as rules allowing to identify identity links across different KGs, possibly described using different ontologies. However, the approach for extracting link keys struggles to scale on large KGs. While embedding-based EM methods efficiently handle large KGs they lack explainability. This paper proposes a novel hybrid EM approach to guarantee the scalability link key extraction approach and improve the explainability of embedding-based EM methods. First, embedding-based EM approaches are used to sample the KGs based on the identity links they generate, thereby reducing the search space to relevant sub-graphs for link key extraction. Second, rules (in the form of link keys) are extracted to explain the generation of identity links by the embedding-based methods. Experimental results demonstrate that the proposed approach allows link key extraction to scale on large KGs, preserving the quality of the extracted link keys. Additionally, it shows that link keys can improve the explainability of the identity links generated by embedding-methods, allowing for the regeneration of 77% of the identity links produced for a specific EM task, thereby providing an approximation of the reasons behind their generation. 3344–3353 10.1145/3696410.3714581

Des agents peuvent faire évoluer leurs ontologies en accomplissant conjointement une tâche. Nous considérons un ensemble de tâches dont chaque agent ne considère qu'une partie. Nous supposons que moins un agent considère de tâches, plus la précision de sa meilleure tâche sera élevée. Pour le vérifier, nous simulons différentes populations considérant un nombre de tâches croissant. De manière contre-intuitive, l'hypothèse n'est pas vérifiée. D'une part, lorsque les agents ont une mémoire illimitée, plus un agent considère de tâches, plus il est précis. D'autre part, lorsque les agents ont une mémoire limitée, les objectifs de maximiser la précision de leur meilleures tâches et de s'accorder entre eux sont mutuellement exclusifs. Lorsque les sociétés favorisent la spécialisation, les agents n'améliorent pas leur précision. Cependant, ces agents décideront plus souvent en fonction de leurs meilleures tâches, améliorant ainsi la performance de leur société. 76–85 9782383950349

14282 978-3-031-43263-7 10.1007/978-3-031-43264-4_28 425–434 Agents have been previously shown to evolve their ontologies while interacting over a single task. However, little is known about how interacting over several tasks affects the accuracy of agent ontologies. Is knowledge learned by tackling one task beneficial for another task? We hypothesize that multi-tasking agents tackling tasks that rely on the same properties, are more accurate than multi-tasking agents tackling tasks that rely on different properties. We test this hypothesis by varying two parameters. The first parameter is the number of tasks assigned to the agents. The second parameter is the number of common properties among these tasks. Results show that when deciding for different tasks relies on the same properties, multi-tasking agents reach higher accuracy. This suggests that when agents tackle several tasks, it is possible to transfer knowledge from one task to another.

Thèse d'informatique 2024GRALM078 This thesis explores how the presence of several tasks affects the cultural evolution of knowledge within agent societies. Previous findings show that when agents coordinate over a single shared task, this improves their accuracy. We build upon these by investigating how coordinating over additional tasks influences the cultural evolution of knowledge, particularly when the cognitive resources of agents are limited. Our goal is to determine whether the cultural evolution of knowledge guarantees: (1) equidistribution of collective resources among different tasks, and (2) improvement of a society’s knowledge across all tasks. To that end, we propose a framework allowing for examining what pushes agents to specialize and how this specialization affects both individual agents and agent societies. Our findings reveal several key insights: First, agents improve their knowledge on one task by undertaking additional tasks. Second, while assigning different tasks to different agents benefits their societies, it is not what triggers agent specialization. On the contrary, specialization emerges when agents allocate limited cognitive resources to these tasks. Third, while temporarily detrimental to their knowledge correctness, forgetting randomly selected knowledge brings long-term benefits when agents are assigned a single task. Lastly, when agents specializing in different tasks interact, one task will gradually monopolize the collectively available resources, causing the entire society to specialize in the same task. These results improve our understanding of how the formation, evolution, and prevalence of specialized knowledge affect individual agents and agent societies.

Echo chamber, the state in which agents are split into groups sharing the same opinions, is a well-known phenomenon in social networks. Opinion dynamics models have been proposed to explain how the phenomenon occurs through agents revising their opinions. However, social network users also exchange beliefs supporting their opinions. This has not been taken into account. In this paper, we extend an existing opinion dynamics model by allowing agents to exchange and update both beliefs and opinions. The process of updating beliefs is described based on the classical belief revision theory. Beliefs and opinions can influence each other guided by values that agents share. We compare opinion propagation with respect to belief influence. Simulation results show that echo chambers are reinforced by interactions between opinions and beliefs. 10.11517/pjsai.JSAI2025.0_3K1IS302 3K1IS302–3K1IS302 2758-7347

Multi-agent opinion dynamics and belief propagation have been studied independently. However, people beliefs may influence their opinions and beliefs should not be too dissonant with opinions. This paper considers how to combine opinions and beliefs so that they preserve the cognitive coherence of agents. For that purpose, it integrates social opinion and belief propagation using classical procedures and two new specific operations guided by values held by agents. In order to assess the effect of this model, we experiment with such agents, varying the processing workflow, graph topology, initial beliefs and opinions, and values. Results show that, by maintaining the coherence between beliefs and opinions of individual agents, the social beliefs and opinions resulting from their propagation is indeed affected. In particular, contrary to what happens with the classical opinion dynamics and belief propagation procedures, connecting opinions and beliefs make them not necessarily converge and not even stabilize. This means that the model supports agents having polarized opinions and beliefs. The outcome of the propagation depends on the workflow, topics, values, and initial graph topology, beliefs and opinions. 10.65109/PQIV4263 1011–1022 2184-433X 978-989-758-796-2

L’ingénierie des connaissances est une thématique de l’Intelligence Artificielle qui contribue au développement de modèles, méthodes et outils pour l’acquisition, la représentation et l’intégration de connaissances. Sa finalité est la production de méthodes et outils « intelligents », capables d’aider l’humain dans ses activités et ses prises de décisions. La conférence Ingénierie des Connaissances est un lieu d’échanges et de réflexions, de présentation et de confrontation des théories, pratiques, méthodes et outils autour de l’ingénierie des connaissances. Cette communauté prend désormais en compte l’essor des algorithmes d’apprentissage automatique et leurs retombées sur les pratiques individuelles et collectives, tout en conservant l’humain au centre des systèmes de décision exploitant les données et les connaissances. Ce numéro spécial regroupe des versions étendues d’une sélection des meilleurs articles des éditions 2021, 2022, et 2023. 6 1–2 1–4 2967-9672 10.5802/ROIA.90FR

10.48448/g9wb-en04

Master's thesis Cultural values are cognitive representations of general objectives, such as independence or mastery, that people use to distinguish whether something is "good" or "bad". More specifically, people may use their values to evaluate alternatives and pick the most compatible one. Cultural values have been previously used in artificial societies of agents with the purpose of replicating and predicting human behavior. However, to the best of our knowledge, they have never been used in the context of cultural knowledge evolution. We consider cooperating agents which adapt their individually learned ontologies by interacting with each other to agree. When two agents disagree during an interaction, one of them needs to adapt its ontology. We use the cultural values of independence, novelty, authority and mastery to influence the choice of which agent adapts in a population of agents sharing the same values. We investigate the effects the choice of cultural values has on the knowledge obtained. Our results show that agents do not improve the accuracy of their knowledge without using the mastery value. Under certain conditions, independence causes the agents to converge to successful interactions faster, and novelty increases knowledge diversity, but both effects come with a large reduction in accuracy. We however did not find any significant effects of authority.

The built environment consists of plenty of physical assets with which we interact on a daily basis. In order to improve not only our built environment, but also our interaction with that environment, we would benefit a lot from semantic representations of this environment. This not only includes buildings, but also large infrastructure (bridges, tunnels, waterways, underground systems), and geospatial data. With this special issue, an insight is given into the current state of the art in terms of semantic technologies and interoperability in this built environment. This editorial not only summarizes the content of the Special Issue on Semantic Technologies and interoperability in the Built Environment, it also provides a brief overview of the current state of the art in general in terms of standardisation and community efforts. 9 6 731–734 10.3233/SW-180321

Linked Data provides access to huge, continuously growing amounts of open data and ontologies in RDF format that describe entities, links and properties on those entities. Equipping Linked Data with inference paves the way to make the Semantic Web a reality. In this survey, we describe a unifying framework for RDF ontologies and databases that we call deductive RDF triplestores. It consists in equipping RDF triplestores with Datalog inference rules. This rule language allows to capture in a uniform manner OWL constraints that are useful in practice, such as property transitivity or symmetry, but also domain-specific rules with practical relevance for users in many domains of interest. The expressivity and the genericity of this framework is illustrated for modeling Linked Data applications and for developing inference algorithms. In particular, we show how it allows to model the problem of data linkage in Linked Data as a reasoning problem on possibly decentralized data. We also explain how it makes possible to efficiently extract expressive modules from Semantic Web ontologies and databases with formal guarantees, whilst effectively controlling their succinctness. Experiments conducted on real-world datasets have demonstrated the feasibility of this approach and its usefulness in practice for data integration and information extraction.

10370 978-3-319-61032-0 121–166 10.1007/978-3-319-61033-7_5

Existing metrics for evaluating complex ontology matching systems often fail to adequately capture the intricacies of (m:n) correspondences. This limitation results in partial or biased alignment quality assessments. This paper introduces a novel metric specifically tailored for complex ontology matching, extending traditional evaluation frameworks by incorporating subgraph similarity measures to ensure structural consistency with reference alignments. It utilizes a tree similarity-based approach, ensuring robustness against common issues such as order variance and detecting incorrect correspondences while adhering to key evaluation properties like completeness and correctness. Empirical experiments conducted on the OAEI complex track datasets demonstrate the superior adaptability of the metric in distinguishing correct structural correspondences compared to conventional and instance-based evaluation methods.

15718 978-3-030-77866-8 10.1007/978-3-031-94575-5_5 77–93

This paper presents CMatch (Complex Matcher), an LLM-based ontology matching approach designed for the Complex Track of the Ontology Alignment Evaluation Initiative (OAEI) 2025. CMatch addresses two core challenges in complex ontology matching: the combinatorial explosion of candidate subgraph pairs and the generation of expressive, logic-based correspondences (e.g., n:m correspondences involving class unions or property compositions). 153–158

Master's thesis In cultural knowledge evolution simulated by multi-agent simulations, agents can improve the accuracy of their knowledge by interacting with other agents and adapting their knowledge with the aim of agreeing. But their knowledge might be confined to specific areas because they do not have the capacity to explore the world around them. Since intrinsic motivation to explore in artificial agents has already proven to increase exploration, it was researched whether and how agents in simulations of cultural knowledge evolution can be motivated to explore. Moreover, it was tested how far this improves and changes their knowledge. Three different kinds of motivation were investigated: curiosity, creativity and non-exploration. Moreover, intrinsic motivation was modelled with and without reinforcement learning. Agents either explored on their own or picked specific interaction partner(s). It has been shown that it is possible to model agents with intrinsic motivation to explore in cultural knowledge evolution, and that this has a significant effect on the agents’ knowledge. Contrary to the expectations and other studies, this did not lead to an increase in knowledge completeness. Out of all intrinsic motivations, curiosity had the highest accuracy and completeness. Models with reinforcement learning performed similar to direct models. As expected, intrinsic motivation led to faster convergence of the agents’ knowledge, especially with social agents. Heterogeneously motivated agents only had a higher accuracy and completeness than homogeneously motivated agents in specific cases. This thesis can be regarded as a foundation for further investigation into the role of intrinsic motivation in cultural knowledge evolution. Different forms of intrinsic motivation or different reinforcement learning techniques could be tested. Additionally, intrinsic motivation at different stages of the experiment or in different ratios, for example curious agents and agents with no motivation, could be investigated in more detail. Lastly, agents could teach other agents things they explored a lot.

Ontology Matching is the task of finding a set of entity correspondences between a pair of ontologies, i.e. an alignment. It has been receiving a lot of attention due to its broad applications. Many techniques have been proposed, among which the ones applying interactive strategies. An interactive ontology matching strategy uses expert knowledge towards improving the quality of the final alignment. When these strategies are based on the expert feedback to validate correspondences, it is important to establish criteria for selecting the set of correspondences to be shown to the expert. A bad definition of this set can prevent the algorithm from finding the right alignment or it can delay convergence. In this work we present techniques which, when used simultaneously, improve the set of candidate correspondences. These techniques are incorporated in an interactive ontology matching approach, called ALINSyn. Experiments successfully show the potential of our proposal. 13–24

Interactive Ontology Matching considers the participation of domain experts during the matching process of two ontologies. An important step of this process is the selection of mappings to submit to the expert. These mappings can be between concepts, attributes or relationships of the ontologies. Existing approaches define the set of mapping suggestions only in the beginning of the process before expert involvement. In previous work, we proposed an approach to refine the set of mapping suggestions after each expert feedback, benefiting from the expert feedback to form a set of mapping suggestions of better quality. In this approach, only concept mappings were considered during the refinement. In this paper, we show a new approach to evaluate the benefit of also considering attribute mappings during the interactive phase of the process. The approach was evaluated using the OAEI conference data set, which showed an increase in recall without sacrificing precision. The approach was compared with the state-of-the-art, showing that the approach has generated alignment with state-of-the-art quality. 25–36

Ontology matching aims at discovering mappings between the entities of two ontologies. It plays an important role in the integration of heterogeneous data sources that are described by ontologies. Interactive ontology matching involves domain experts in the matching process. In some approaches, the expert provides feedback about mappings between ontology entities, i.e., these approaches select mappings to present to the expert who replies which of them should be accepted or rejected, so taking advantage of the knowledge of domain experts towards finding an alignment. In this paper, we present Alin, an interactive ontology matching approach which uses expert feedback not only to approve or reject selected mappings, but also to dynamically improve the set of selected mappings, i.e., to interactively include and to exclude mappings from it. This additional use for expert answers aims at increasing in the benefit brought by each expert answer. For this purpose, Alin uses four techniques. Two techniques were used in previous versions of Alin to dynamically select concept and attribute mappings. Two new techniques are introduced in this paper: one to dynamically select relationship mappings and another one to dynamically reject inconsistent selected mappings using anti-patterns. We compared Alin with state-of-the-art tools, showing that it generates alignment of comparable quality. 35 e1 10.1017/S0269888919000249

Master's thesis Vast amounts of RDF data are made available on the web by various institutions providing overlapping information. To be fully exploited, different representations of the same object across various data sets have to be identified. This is what is called data interlinking. One novel way to generate such links is to use link keys. Link keys generalise database keys by applying them across two data sets. The structure of RDF makes this problem much more complex than for relational databases for several reasons. An instance can have multiple values for a given attribute. Moreover, values of properties are not necessarily datatypes but instances of the graph. A first method has been designed to extract and select link keys from two classes of objects which deals with multiple values but not object values. Moreover, the extraction step has been rephrased in formal concept analysis (FCA) allowing to generate link keys across relational tables. Our aim is to extend this work so that it can deal with multiple values. Then, we show how to use it to deal with object values when the data set is cycle free. This encoding does not necessarily generate the optimal link keys. Hence, we use relational concept analysis (RCA), an extension of FCA taking relations between concepts into account. We show that a new expression of this problem is able to extract the optimal link keys even in the presence of circularities. Moreover, the elaborated process does not require information about the alignments of the ontologies to find out for which pairs of classes to extract link keys. We implemented these methods and evaluated them by reproducing the experiments made in previous studies. This shows that the method extracts the expected results as well as (also expected) scalability issues.

Ontology alignments enable interoperability between heterogeneous information resources. The Alignment Repair Game (ARG) specifically provides a way for agents to simultaneously communicate and improve the alignment when a communication failure occurs. This is achieved through applying adaptation operators that provide a revision strategy for agents to resolve failures with minimum information loss. In this paper, we explore how closely these operators resemble logical dynamics. We develop a variant of Dynamic Epistemic Logic called DEOL to capture the dynamics of ARG by modeling ontologies as knowledge and alignments as belief with respect to the plausibility relation. The dynamics of ARG are then achieved through announcements and conservative upgrades. With the representation of ARG in DEOL, we formally establish the limitations and the redundancy of the adaptation operators. More precisely, that for a complete logical reasoner, replace, addjoin and refine are redundant for one or both agents in the game and that add would be replaced by addjoin in all cases.

Ontology alignments enable agents to communicate while preserving heterogeneity in their information. Alignments may not be provided as input and should be able to evolve when communication fails or when new information contradicting the alignment is acquired. In the Alignment Repair Game (ARG) this evolution is achieved via adaptation operators. ARG was evaluated experimentally and the experiments showed that agents converge towards successful communication and improve their alignments. However, whether the adaptation operators are formally correct, complete or redundant is still an open question. In this paper, we introduce a formal framework based on Dynamic Epistemic Logic that allows us to answer this question. This framework allows us (1) to express the ontologies and alignments used, (2) to model the ARG adaptation operators through announcements and conservative upgrades and (3) to formally establish the correctness, partial redundancy and incompleteness of the adaptation operators in ARG. 1422–1430 978-1-4503-7518-4 2523-5699

Public signature awareness is satisfied if agents are aware of the vocabulary, propositions, used by other agents to think and talk about the world. However, assuming that agents are fully aware of each other's signatures prevents them to adapt their vocabularies to newly gained information, from the environment or learned through agent communication. Therefore this is not realistic for open multi-agent systems. We propose a novel way to model awareness with partial valuations that drops public signature awareness and can model agent signature unawareness, and we give a first view on defining the dynamics of raising and forgetting awareness on this framework.

Gossip describes the spread of information throughout a network of agents. It investigates how agents, each starting with a unique secret, can efficiently make peer-to-peer calls so that ultimately everyone knows all secrets. In Dynamic Gossip, agents share phone numbers in addition to secrets, which allows the network to grow at run-time. Most gossip protocols assume that all agents are reliable, but this is not given for many practical applications. We drop this assumption and study Dynamic Gossip with unreliable agents. The aim is then for agents to learn all secrets of the reliable agents and to identify the unreliable agents. We show that with unreliable agents classic results on Dynamic Gossip no longer hold. Specifically, the Learn New Secrets protocol is no longer characterised by the same class of graphs, so-called sun graphs. In addition, we show that unreliable agents that do not initiate communication are harder to identify than agents that do. This has paradoxical consequences for measures against unreliability, for example to combat the spread of fake news in social networks. 51–67 10.1007/978-3-030-65840-3_4

12569 978-3-030-65839-7 978-3-030-65840-3

Partial Dynamic Epistemic Logic allows agents to have different knowledge representations about the world through agent awareness. Agents use their own vocabularies to reason and talk about the world and raise their awareness when confronted with new vocabulary. Through raising awareness the vocabularies of agents are extended, suggesting there is a dual, inverse operator for forgetting awareness that decreases vocabularies. In this paper, we discuss such an operator. Unlike raising awareness, this operator may induce an abstraction on models that removes evidence while preserving conclusions. This is useful to better understand how agents with different knowledge representations communicate with each other, as they may forget the justifications that led them to their conclusions. 18–21

Ontology alignments enable agents to communicate while preserving heterogeneity in their knowledge. Alignments may not be provided as input and should be able to evolve when communication fails or when new information contradicting the alignment is acquired. The Alignment Repair Game (ARG) has been proposed for agents to simultaneously communicate and repair their alignments through adaptation operators when communication failures occur. ARG has been evaluated experimentally and the experiments showed that agents converge towards successful communication and improve their alignments. However, whether the adaptation operators are formally correct, complete or redundant could not be established by experiments. We introduce a logical model, Dynamic Epistemic Ontology Logic (DEOL), that enables us to answer these questions. This framework allows us (1) to express the ontologies and alignments used via a faithful translation from ARG to DEOL, (2) to model the ARG adaptation operators as dynamic modalities and (3) to formally define and establish the correctness, partial redundancy and incompleteness of the adaptation operators in ARG. 35 2 32 10.1007/s10458-021-09508-8

Thèse de mathématiques-informatique 2021GRALM045 To reason and talk about the world, agents may use their own distinct vocabularies, structured into knowledge representations, also called ontologies. In order to communicate, they use alignments: translations between terms of their ontologies. aHowever, ontologies may change, requiring their alignments to evolve accordingly. Experimental cultural evolution offers a framework to study the mechanisms of their knowledge evolution. It has been applied to the evolution of alignments in the Alignment Repair Game (ARG). Experiments have shown that, through ARG, agents improve their alignments and reach successful communication. Yet, these experiments are not sufficient to understand the formal properties of cultural knowledge evolution. This thesis bridges experimental cultural knowledge evolution with a theoretical model of cultural knowledge evolution in logic. This is achieved by introducing Dynamic Epistemic Ontology Logic and defining a faithful translation of ARG in DEOL that (a) encodes the ontologies, (b) maps agents' ontologies and alignments to knowledge and beliefs, and (c) captures the adaptation operators through announcements and conservative upgrades. This model shows that all but one adaptation operator are correct, they are incomplete and some are partially redundant. Three differences between the ARG agents and their logical model explain these results, leading to an independent model of awareness based on partial valuations and weakly reflexive relations. An alternative model of ARG is then defined under which the formal properties are re-examined, showing that this model is closer to the original game. This is a first step towards defining a theoretical model of cultural knowledge evolution.

Pedagogical material Class? is an enjoyable card game aiming at grouping colourful cards into meaningful classes. It illustrates facets of reasoning with classifications. In order to introduce Class? progressively, this small gamebook provides a sequence of games before getting to the Class? game itself and beyond. The games are presented in increasing order of difficulty so that a game will benefit from mastering of previous ones.

Agents use their own vocabularies to reason and talk about the world. Public signature awareness is satisfied if agents are aware of the vocabularies, or signatures, used by all agents they may, eventually, interact with. Multi-agent modal logics and in particular Dynamic Epistemic Logic rely on public signature awareness for modeling information flow in multi-agent systems. However, this assumption is not desirable for dynamic and open multi-agent systems because (1) it prevents agents to use unique signatures other agents are unaware of, (2) it prevents agents to openly extend their signatures when encountering new information, and (3) it requires that all future knowledge and beliefs of agents are bounded by the current state. We propose a new semantics for awareness that enables us to drop public signature awareness. This semantics is based on partial valuation functions and weakly reflexive relations. Dynamics for raising public and private awareness are then defined in such a way as to differentiate between becoming aware of a proposition and learning its truth value. With this, we show that knowledge and beliefs are not affected through the raising operations. 91 4 431–464 10.1007/s10472-022-09809-y

Class? est un jeu dans lequel les joueurs doivent organiser leurs cartes en fonction d’une classification cachée. Les cartes posées par les autres joueurs leur permettent de deviner où mettre les leurs. Il a été conçu pour que des écoliers appréhendent que les mêmes objets peuvent être classés de différentes manières et qu’il est possible de transmettre une classification sans l’expliciter. Le jeu fait appel a des notions qui se présentent facilement à l’aide des cartes à jouer comme des ensembles définis par des conditions nécessaires et suffisantes (classes). Cela permet d’introduire des classifications hiérarchiques et des notions algorithmiques (tests de conditions, récursion) pour les manipuler. Enfin, il nécessite de raisonner logiquement sur ces notions. Class? a été joué avec succès par des élèves du CM2 à la seconde. Nous nous sommes donc posé la question de son positionnement en tant que ressource pédagogique. Il apparait tout d’abord qu’il ne semble pas illustrer de concepts particulièrement mis en avant par les programmes officiels de l’éducation nationale. Il offre plutôt une manière alternative de renforcer des notions transversales très importantes en informatique. Nous caractérisons Class? par rapport aux efforts d’informatique sans ordinateurs et à d’autres jeux utilisables à cette fin. Finalement, nous discutons d’une décomposition de Class? en une succession de jeux plus simples permettant d’introduire les notions impliquées l’une après l’autre. 1 1 10.52497/radix.v1i1.401

Reproducibility is a desirable property of scientific research. On the one hand, it increases confidence in results. On the other hand, reproducible results can be extended on a solid basis. In rapidly developing fields such as machine learning, the latter is particularly important to ensure the reliability of research. In this paper, we present a systematic approach to reproducing (using the available implementation), replicating (using an alternative implementation) and reevaluating (using different datasets) state-of-the-art experiments. This approach enables the early detection and correction of deficiencies and thus the development of more robust and transparent machine learning methods. We detail the independent reproduction, replication, and reevaluation of the initially published experiments with a method that we want to extend. For each step, we identify issues and draw lessons learned. We further discuss solutions that have proven effective in overcoming the encountered problems. This work can serve as a guide for further reproducibility studies and generally improve reproducibility in machine learning. 978-1-57735-887-9 10.1609/aaai.v38i14.29515 15850–15858

Query containment is defined as the problem of determining if the result of a query is included in the result of another query for any dataset. It has major applications in query optimization and knowledge base verification. The main objective of this work is to provide sound and complete procedures to determine containment of SPARQL queries under expressive description logic schema axioms. Beyond that, these procedures are experimentally evaluated. To date, testing query containment has been performed using different techniques: containment mapping, canonical databases, automata theory techniques and through a reduction to the validity problem in logic. In this work, we use the latter technique to test containment of SPARQL queries using an expressive modal logic called mu-calculus. For that purpose, we define an RDF graph encoding as a transition system which preserves its characteristics. In addition, queries and schema axioms are encoded as mu-calculus formulae. Thereby, query containment can be reduced to testing validity in the logic. We identify various fragments of SPARQL and description logic schema languages for which containment is decidable. Additionally, we provide theoretically and experimentally proven procedures to check containment of these decidable fragments. Finally, we propose a benchmark for containment solvers which is used to test and compare the current state-of-the-art containment solvers. 7 3 133–154 10.1007/s13740-018-0087-1