TY - JOUR TI - Project VoLL-KI : Learning from Learners AU - Kohlhase, Michael AU - Berges, Marc AU - Grubert, Jens AU - Landes, Dieter AU - Leidner, Jochen L. AU - Henrich, Andreas AU - Mittag, Florian AU - Nicklas, Daniela AU - Schmid, Ute Maria AU - Sedlmaier, Yvonne AU - Ulbrich-vom Ende, Achim AU - Wolter, Diedrich PY - 2026 VL - 39 SP - 299–309 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/115195 LA - eng AB - “Learning from Learners” (“Von Lernenden Lernen”, “VoLL-KI ” for short) is a is collaborative research project with the goal of creating a practical toolbox of instruments at diferent levels of abstraction to improve the learning experience and outcomes for students of artifcial intelligence. Learning and teaching at tertiary education institutions are currently ITbased but not AI-supported. The VoLL-KI project aims to go this crucial next step. Using AI approaches (e.g., ML, symbolic AI, statistical AI), several educational technologies are developed on the diferent granularity of the study programs (e.g., course, semester, program). So, dashboards for study planning, recommender systems improving student advisory services and learning material selection, VR-based learning experiences, educational chatbots, and adaptive learning environments are developed. The systems are mostly deployed as prototypes KW - Computer-aided education KW - Intelligent tutoring systems KW - Teaching support for artifcial intelligence education KW - Computer-aided learning (CAL) KW - Data-driven learning quality management KW - Study course monitoring KW - Learning analytics ER - TY - RPRT TI - Behind The Corner : Using Qualitative Reasoning for Solving Angry Birds AU - Haase, Felix AU - Wolter, Diedrich PY - 2026 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/114349 LA - eng AB - Physics-based logic puzzle games present diverse challenges for AI, primarily related to coping with uncertainty in complex continuous environments. As these challenges tie in with the goals of qualitative reasoning (QR), in particular to plan, predict and diagnose physical mechanisms, these games pose a natural benchmark for QR. In this paper we consider the projectile collision games Angry Birds by Rovio Entertainment on which the AI bench- mark AIBirds is based. We investigate whether a qualitative approach to action planning under uncertainty proposed by Ge et al. (2016) can be adapted to the domain of Angry Birds in order to identify targets hittable by multiple rebounds. It turns out that the search space for solving Angry Birds is complexly structured and a fine-grained decomposition is required, leading to high computational costs. This paper presents an analysis of the problem and discusses means to overcome the problems faced. We propose an improved qualitative prediction method that is able to solve hard levels, yet it also suggests the need of complementary methods. KW - physical reasoning KW - qualitative simulation KW - action planning ER - TY - JOUR TI - Embedding Ontologies in the Description Logic ALC by Axis-Aligned Cones AU - Lütfü Özçep, Özgür AU - Leemhuis, Mena AU - Wolter, Diedrich PY - 2025 SP - 217-267 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/112045 LA - eng AB - This paper is concerned with knowledge graph embedding with background knowledge, taking the formal perspective of logics. In knowledge graph embedding, knowledge— expressed as a set of triples of the form (a R b) (“a is R-related to b”)—is embedded into a real-valued vector space. The embedding helps exploiting geometrical regularities of the space in order to tackle typical inductive tasks of machine learning such as link prediction. Recent embedding approaches also consider incorporating background knowledge, in which the intended meanings of the symbols a, R, b are further constrained via axioms of a theory. Of particular interest are theories expressed in a formal language with a neat semantics and a good balance between expressivity and feasibility. In that case, the knowledge graph together with the background can be considered to be an ontology. This paper develops a cone-based theory for embedding in order to advance the expressivity of the ontology: it works (at least) with ontologies expressed in the description logic ALC, which comprises restricted existential and universal quantifers, as well as concept negation and concept disjunction. In order to align the classical Tarskian Style semantics for ALC with the sub-symbolic representation of triples, we use the notion of a geometric model of an ALC ontology and show, as one of our main results, that an ALC ontology is satisfable in the classical sense if it is satisfable by a geometric model based on cones. The geometric model, if treated as a partial model, can even be chosen to be faithful, i.e., to refect all and only the knowledge captured by the ontology. We introduce the class of axis-aligned cones and show that modulo simple geometric operations any distributive logic (such as ALC) interpreted over cones employs this class of cones. Cones are also attractive from a machine learning perspective on knowledge graph embeddings since they give rise to applying conic optimization techniques. KW - - ER - TY - JOUR TI - On domain generators for the evaluation of action reversibility in STRIPS AU - Schwartz, Tobias AU - Boockmann, Jan H. AU - Martin, Leon PY - 2025 SP - 699-726 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/111383 LA - eng AB - Robustness is a crucial requirement for the deployment of AI systems in real-world scenarios. In the context of AI planning, the concept of action reversibility, i.e., the ability to undo the effects of an action using a reverse plan, is a promising direction for achieving robust plans. Plans composed exclusively of reversible actions exhibit resilience against goal changes during the execution of the plan. However, the evaluation of action reversibility systems in STRIPS planning presents a challenge, given that standard planning benchmarks are often not suitable. Early experiments using a naive implementation of an action reversibility algorithm show that the available domain generation approach is susceptible to bias. Building on this existing domain generator, we introduce two slight variations that exhibit entirely different search space characteristics. We assess these domain generators using the naive action reversibility implementation and existing ASP implementations, and demonstrate that different generators indeed favor different implementations. As a follow-up to this line of research, we present a generalized domain generator facilitating the creation of domains with diverse search space characteristics. To finally reduce the utilization of contrived generation patterns, we propose another domain generator based on the Barabási-Albert model yielding less rigid domains. Our experiments demonstrate that these new domain generators can produce a variety of domains with diverse search space characteristics, enabling a less biased evaluation of action reversibility systems. KW - STRIPS planning KW - Action reversibility KW - Benchmark KW - PDDL domain generation ER - TY - CPAPER TI - Georeferencing River Networks Using Spatial Reasoning AU - Wolter, Diedrich AU - Blank, Daniel AU - Henrich, Andreas PY - 2025 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/106525 LA - eng AB - This paper is concerned with automatic georeferencing of river networks from raster images such as aerial photos or maps. Determining reasonable assignments between a given network of rivers derived from a textual description and an image is subject to high combinatorial complexity and uncertainty. We investigate the application of spatial reasoning in automatic georeferencing. KW - georeferencing river networks KW - spatial reasoning KW - GIS ER - TY - CPAPER TI - Diagnosing Algorithms by Abduction from Manual Simulation AU - Bayerkuhnlein, Moritz AU - Wolter, Diedrich PY - 2025 T2 - Proceedings of the Second Work shop on Artificial Intelligence for Artificial Intelligence Education (AI4AI Learning 2024) SP - 49-58 PB - University of Bamberg Press CY - Bamberg SN - 978-3-98989-054-1 DO - 10.20378/irb-108887 UR - https://fis.uni-bamberg.de/handle/uniba/108887 LA - eng AB - Taking a conceptual idea and turning it into a precise algorithm is at the heart of computational thinking. However, novice programmers often struggle when their code does not behave as they intended. This paper identifies problems in pseudocode algorithms based on deviations from observations that could be gathered during manual simulations by the programmer. By applying model-based diagnosis to the faulty pseudocode, informed by manual simulation, we locate errors and suggest fixes. The diagnosis results in the identification of a specific location in the algorithm and provides an output description for the faulty part that matches the programmer's intent during the manual simulation, thus aiding in debugging. KW - - ER - TY - CPAPER TI - Proceedings of the Second Work shop on Artificial Intelligence for Artificial Intelligence Education (AI4AI Learning 2024) AU - Singla, Adish AU - Hobert, Sebastian AU - Joachim, Silvia AU - Hennecke, Martin AU - Voelker, Adrian AU - Thaler, Anna Magdalena AU - Summerer, Maximilian T. AU - Schmid, Ute AU - Bayerkuhnlein, Moritz AU - Wolter, Diedrich AU - Bugert, Flemming AU - Bittner, Dominik AU - Ezer, Timur AU - Nadimpalli, Vamsi Krishna AU - Staufer, Susanne AU - Hauser, Florian AU - Maier, Robert AU - Grabinger, Lisa AU - Mottok, Jürgen AU - Niemann, Sonja ED - Schmid, Ute Maria ED - Leidner, Jochen L. ED - Kohlhase, Michael ED - Wolter, Diedrich PY - 2025 PB - University of Bamberg Press CY - Bamberg T3 - Schriften aus der Fakultät Wirtschaftsinformatik und Angewandte Informatik der Otto-Friedrich-Universität Bamberg SN - 978-3-98989-054-1 DO - 10.20378/irb-107661 UR - https://fis.uni-bamberg.de/handle/uniba/107661 LA - eng AB - Artificial Intelligence (AI) methods for education have been a topic of AI research since many decades with focus on Intelligent Tutoring Systems (ITS). ITS typically address a specific knowledge domain with a focus on individual learning support based on learning by problem solving, identification of misconceptions, and tailored feedback. Methods often are a combination of knowledge-based and machine learning approaches. Since the emergence of technologies based on Large Language Models (LLMs), the use of LLMs in educational contexts has received much attention, resulting in novel approaches. The Second Workshop on Artificial Intelligence for Artificial Intelligence Education (AI4AILearning) has been held on September 24, 2024, in conjunction with the 47th German Conference on Artificial Intelligence in Würzburg, Germany. The workshop has a focus on research of AI methods for teaching AI competencies including programming skills. It provides a platform for exchange of ideas and experiences under the general theme of AI for education, specialising on university education in AI. The workshop includes interdisciplinary contributions from cognitive science and education technology. KW - AI Education KW - Programming Education KW - Technology-Enhanced-Learning KW - Intelligent Tutoring Systems KW - Large Language Models ER - TY - JOUR TI - On domain generators for the evaluation of action reversibility in STRIPS AU - Schwartz, Tobias AU - Boockmann, Jan H. AU - Martin, Leon PY - 2025 JO - Annals of Mathematics and Artificial Intelligence VL - 93 IS - 5 SP - 699-726 PB - Springer Science + Business Media B.V CY - Dordrecht [u.a.] SN - 1573-7470 DO - 10.1007/s10472-024-09960-8 UR - https://fis.uni-bamberg.de/handle/uniba/107589 LA - eng AB - Robustness is a crucial requirement for the deployment of AI systems in real-world scenarios. In the context of AI planning, the concept of action reversibility, i.e., the ability to undo the effects of an action using a reverse plan, is a promising direction for achieving robust plans. Plans composed exclusively of reversible actions exhibit resilience against goal changes during the execution of the plan. However, the evaluation of action reversibility systems in STRIPS planning presents a challenge, given that standard planning benchmarks are often not suitable. Early experiments using a naive implementation of an action reversibility algorithm show that the available domain generation approach is susceptible to bias. Building on this existing domain generator, we introduce two slight variations that exhibit entirely different search space characteristics. We assess these domain generators using the naive action reversibility implementation and existing ASP implementations, and demonstrate that different generators indeed favor different implementations. As a follow-up to this line of research, we present a generalized domain generator facilitating the creation of domains with diverse search space characteristics. To finally reduce the utilization of contrived generation patterns, we propose another domain generator based on the Barabási-Albert model yielding less rigid domains. Our experiments demonstrate that these new domain generators can produce a variety of domains with diverse search space characteristics, enabling a less biased evaluation of action reversibility systems. KW - STRIPS planning KW - Action reversibility KW - Benchmark KW - PDDL domain generation ER - TY - JOUR TI - Project VoLL-KI : Learning from Learners AU - Kohlhase, Michael AU - Berges, Marc AU - Grubert, Jens AU - Landes, Dieter AU - Leidner, Jochen L. AU - Henrich, Andreas AU - Mittag, Florian AU - Nicklas, Daniela AU - Schmid, Ute AU - Sedlmaier, Yvonne AU - Ulbrich-vom Ende, Achim AU - Wolter, Diedrich PY - 2025 JO - Künstliche Intelligenz : KI ; Forschung, Entwicklung, Erfahrungen ; Organ des Fachbereichs 1 Künstliche Intelligenz der Gesellschaft für Informatik e.V., GI VL - 39 SP - 299–309 PB - Springer CY - Berlin ; Heidelberg SN - 0933-1875 DO - 10.1007/s13218-024-00846-9 UR - https://fis.uni-bamberg.de/handle/uniba/107128 LA - eng AB - “Learning from Learners” (“Von Lernenden Lernen”, “VoLL-KI ” for short) is a is collaborative research project with the goal of creating a practical toolbox of instruments at diferent levels of abstraction to improve the learning experience and outcomes for students of artifcial intelligence. Learning and teaching at tertiary education institutions are currently ITbased but not AI-supported. The VoLL-KI project aims to go this crucial next step. Using AI approaches (e.g., ML, symbolic AI, statistical AI), several educational technologies are developed on the diferent granularity of the study programs (e.g., course, semester, program). So, dashboards for study planning, recommender systems improving student advisory services and learning material selection, VR-based learning experiences, educational chatbots, and adaptive learning environments are developed. The systems are mostly deployed as prototypes KW - Computer-aided education KW - Intelligent tutoring systems KW - Teaching support for artifcial intelligence education KW - Computer-aided learning (CAL) KW - Data-driven learning quality management KW - Study course monitoring KW - Learning analytics ER - TY - CHAP TI - Model-Based-Diagnosis for Assistance in Programming Exercises AU - Bayerkuhnlein, Moritz AU - Wolter, Diedrich PY - 2024 T2 - Artificial Intelligence : ECAI 2023 International Workshops ; XAI^3, TACTIFUL, XI-ML, SEDAMI, RAAIT, AI4S, HYDRA, AI4AI ; Kraków, Poland, September 30 – October 4, 2023 ; Proceedings, Part 2 SP - 459-470 PB - Springer CY - Cham, Switzerland SN - 978-3-031-50484-6 DO - 10.1007/978-3-031-50485-3_44 UR - https://fis.uni-bamberg.de/handle/uniba/112332 LA - eng ER - TY - CPAPER TI - From Resolving Inconsistencies in Qualitative Constraints Networks to Identifying Robust Solutions : A Universal Encoding in ASP AU - Bayerkuhnlein, Moritz AU - Schwartz, Tobias AU - Wolter, Diedrich PY - 2024 T2 - KI 2024: Advances in Artificial Intelligence : 47th German Conference on AI Würzburg, Germany, September 25–27, 2024, Proceedings SP - 3-16 PB - Springer CY - Cham, Switzerland T3 - Lecture Notes in Computer Science SN - 978-3-031-70893-0 DO - 10.1007/978-3-031-70893-0_1 UR - https://fis.uni-bamberg.de/handle/uniba/112192 LA - eng ER - TY - CPAPER TI - Model-Based Diagnosis with ASP for Non-groundable Domains AU - Bayerkuhnlein, Moritz AU - Wolter, Diedrich PY - 2024 T2 - Foundations of Information and Knowledge Systems : 13th International Symposium, FoIKS 2024, Sheffield, UK, April 8-11, 2024 ; Proceedings SP - 363-380 PB - Springer CY - Berlin ; Heidelberg T3 - Lecture Notes in Computer Science SN - 978-3-031-56940-1 DO - 10.1007/978-3-031-56940-1_20 UR - https://fis.uni-bamberg.de/handle/uniba/108438 LA - eng ER - TY - JOUR TI - On robust vs fast solving of qualitative constraints AU - Wehner, Jan AU - Sioutis, Michael AU - Wolter, Diedrich PY - 2023 JO - Journal of heuristics VL - 29 IS - 4-6 SP - 461-485 PB - Springer Science + Business Media B.V. CY - Dordrecht [u.a.] SN - 1572-9397 DO - 10.1007/s10732-023-09517-8 UR - https://fis.uni-bamberg.de/handle/uniba/112340 LA - eng ER - TY - JOUR TI - Embedding Ontologies in the Description Logic ALC by Axis-Aligned Cones AU - Lütfü Özçep, Özgür AU - Leemhuis, Mena AU - Wolter, Diedrich PY - 2023 JO - Journal of artificial intelligence research : JAIR ; an international electronic and print journal VL - 78 SP - 217-267 PB - AI Access Found. SN - 1943-5037 DO - 10.1613/jair.1.13939 UR - https://fis.uni-bamberg.de/handle/uniba/109320 LA - eng AB - This paper is concerned with knowledge graph embedding with background knowledge, taking the formal perspective of logics. In knowledge graph embedding, knowledge— expressed as a set of triples of the form (a R b) (“a is R-related to b”)—is embedded into a real-valued vector space. The embedding helps exploiting geometrical regularities of the space in order to tackle typical inductive tasks of machine learning such as link prediction. Recent embedding approaches also consider incorporating background knowledge, in which the intended meanings of the symbols a, R, b are further constrained via axioms of a theory. Of particular interest are theories expressed in a formal language with a neat semantics and a good balance between expressivity and feasibility. In that case, the knowledge graph together with the background can be considered to be an ontology. This paper develops a cone-based theory for embedding in order to advance the expressivity of the ontology: it works (at least) with ontologies expressed in the description logic ALC, which comprises restricted existential and universal quantifers, as well as concept negation and concept disjunction. In order to align the classical Tarskian Style semantics for ALC with the sub-symbolic representation of triples, we use the notion of a geometric model of an ALC ontology and show, as one of our main results, that an ALC ontology is satisfable in the classical sense if it is satisfable by a geometric model based on cones. The geometric model, if treated as a partial model, can even be chosen to be faithful, i.e., to refect all and only the knowledge captured by the ontology. We introduce the class of axis-aligned cones and show that modulo simple geometric operations any distributive logic (such as ALC) interpreted over cones employs this class of cones. Cones are also attractive from a machine learning perspective on knowledge graph embeddings since they give rise to applying conic optimization techniques. KW - - ER - TY - CPAPER TI - Computing Most Likely Scenarios of Qualitative Constraint Networks AU - Schwartz, Tobias AU - Wolter, Diedrich PY - 2023 T2 - KI 2023: Advances in Artificial Intelligence : 46th German Conference on AI, Berlin, Germany, September 26-29, 2023, Proceedings SP - 184-197 PB - Springer Nature Switzerland CY - Cham T3 - Lecture Notes in Computer Science SN - 978-3-031-42607-0 DO - 10.1007/978-3-031-42608-7_15 UR - https://fis.uni-bamberg.de/handle/uniba/91837 LA - eng ER - TY - JOUR TI - Learning with cone-based geometric models and orthologics AU - Leemhuis, Mena AU - Özgür, L. Özçep AU - Wolter, Diedrich PY - 2023 SP - 1159-1195 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/91606 LA - eng AB - Recent approaches for knowledge-graph embeddings aim at connecting quantitative data structures used in machine learning to the qualitative structures of logics. Such embeddings are of a hybrid nature, they are data models that also exhibit conceptual structures inherent to logics. One motivation to investigate embeddings is to design conceptually adequate machine learning (ML) algorithms that learn or incorporate ontologies expressed in some logic. This paper investigates a new approach to embedding ontologies into geometric models that interpret concepts by geometrical structures based on convex cones. The ontologies are assumed to be represented in an orthologic, a logic with a full (ortho)negation. As a proof of concept this cone-based embedding was implemented within two ML algorithms for weak supervised multi-label learning. Both algorithms rely on cones but the first addresses ontologies expressed in classical propositional logic whereas the second addresses a weaker propositional logic, namely a weak orthologic that does not fulfil distributivity. The algorithms were evaluated and showed promising results that call for investigating other (sub)classes of cones and developing fine-tuned algorithms based on them. KW - Concept learning KW - Knowledge graph embedding KW - Support-vector machine KW - Multi-label learning KW - Orthologic ER - TY - JOUR TI - Automated Interpretation of Place Descriptions: Determining Entity Types for Querying OSM AU - Yousaf, Madiha AU - Schwartz, Tobias AU - Wolter, Diedrich PY - 2023 SP - 69-81 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/59712 LA - eng AB - This paper is concerned with interpretation of natural language place descriptions, as they are a rich source of geographic information. A place description is interpreted by matching geographic entities occurring in the text against the OpenStreetMap (OSM) database. This paper is mainly concerned with interpretation of paraphrased places, i.e., entities for which no name is given and which may only by described. Our objective is to determine suitable entity types that allow querying the OpenStreetMap database for the respective place. For example, if we wish to identify a place to eat, we have to check for entities of an a-priori unknown type (cafe, restaurant, etc.). Challenges arise from the open-endedness of language, its ambiguity, and context-sensitivity as well as from mismatches between human conceptualization of place and database ontologies. The contributions of this paper are, first, to present a hard problem that is key to geo-information retrieval beyond named entities. Second, we propose context-sensitive methods for identifying place types based on semantic word similarity. We evaluate the methods on text extracted from Wikipedia and travel blogs, revealing their contribution to advancing automated interpretation of place descriptions to paraphrased places. KW - Geographic information retrieval KW - OSM Tags KW - Volunteered geographic information KW - Place descriptions KW - GIS ER - TY - THES TI - Representing and Reasoning with Context-Sensitive Vague Place Descriptions AU - Yousaf, Madiha PY - 2023 PB - Otto-Friedrich-Universität CY - Bamberg DO - 10.20378/irb-59175 UR - https://fis.uni-bamberg.de/handle/uniba/59175 LA - eng AB - This doctoral thesis aims to design a geographic information system based on locations to make the information about the human location interpretable for computers. This research employs natural language place descriptions as its primary research input and focuses on five research components as main goals towards understanding place descriptions: incorporating context-sensitive information, spatial information extraction, and modeling, reasoning, place geo-referencing, and querying. KW - Artificial Intelligence, Künstliche Intelligenz KW - place description ER - TY - JOUR TI - Automated Interpretation of Place Descriptions: Determining Entity Types for Querying OSM AU - Schwartz, Tobias AU - Wolter, Diedrich AU - Yousaf, Madiha PY - 2023 JO - KI - Künstliche Intelligenz VL - 37 IS - 1 SP - 69-81 PB - Springer CY - Heidelberg SN - 1610-1987 DO - 10.1007/s13218-022-00798-y UR - https://fis.uni-bamberg.de/handle/uniba/59676 LA - eng AB - This paper is concerned with interpretation of natural language place descriptions, as they are a rich source of geographic information. A place description is interpreted by matching geographic entities occurring in the text against the OpenStreetMap (OSM) database. This paper is mainly concerned with interpretation of paraphrased places, i.e., entities for which no name is given and which may only by described. Our objective is to determine suitable entity types that allow querying the OpenStreetMap database for the respective place. For example, if we wish to identify a place to eat, we have to check for entities of an a-priori unknown type (cafe, restaurant, etc.). Challenges arise from the open-endedness of language, its ambiguity, and context-sensitivity as well as from mismatches between human conceptualization of place and database ontologies. The contributions of this paper are, first, to present a hard problem that is key to geo-information retrieval beyond named entities. Second, we propose context-sensitive methods for identifying place types based on semantic word similarity. We evaluate the methods on text extracted from Wikipedia and travel blogs, revealing their contribution to advancing automated interpretation of place descriptions to paraphrased places. KW - Geographic information retrieval KW - OSM Tags KW - Volunteered geographic information KW - Place descriptions KW - GIS ER - TY - CPAPER TI - A Dimensionality Reduction and Reconstruction Method for Data with Multiple Connected Components AU - Yao, Yuqin AU - Gao, Yang AU - Long, Zhiguo AU - Meng, Hua AU - Sioutis, Michael PY - 2022 T2 - 2022 IEEE 5th International Conference on Big Data and Artificial Intelligence (BDAI), proceedings SP - 87-92 PB - IEEE DO - 10.1109/bdai56143.2022.9862787 UR - https://fis.uni-bamberg.de/handle/uniba/115750 LA - eng ER - TY - CPAPER TI - Hybrid AI Systems Grounded on Qualitative Spatio-Temporal Reasoning AU - Sioutis, Michael AU - Long, Zhiguo PY - 2022 T2 - Proceedings of the 12th Hellenic Conference on Artificial Intelligence PB - ACM CY - New York,NY, United States SN - 978-1-4503-9597-7 DO - 10.1145/3549737.3549807 UR - https://fis.uni-bamberg.de/handle/uniba/115751 LA - eng ER - TY - CPAPER TI - Knowledge Graph Embeddings with Ontologies : Reification for Representing Arbitrary Relations AU - Leemhuis, Mena AU - Özçep, Özgür L. AU - Wolter, Diedrich PY - 2022 T2 - KI 2022 : Advances in Artificial Intelligence ; 45th German Conference on AI, Trier, Germany, September 19–23, 2022, Proceedings SP - 146-159 PB - Springer CY - Cham, Switzerland T3 - Lecture Notes in Computer Science SN - 978-3-031-15790-5 DO - 10.1007/978-3-031-15791-2_13 UR - https://fis.uni-bamberg.de/handle/uniba/112632 LA - eng ER - TY - JOUR TI - Approximating Region Boundaries Based on Qualitative and Quantitative Information AU - Long, Zhiguo AU - Sioutis, Michael AU - Li, Qingqian AU - Meng, Hua AU - Li, Heng-Chao PY - 2022 JO - IEEE intelligent systems VL - 37 IS - 6 SP - 38-47 PB - IEEE CY - New York, NY [u.a.] SN - 1541-1672 DO - 10.1109/mis.2022.3204561 UR - https://fis.uni-bamberg.de/handle/uniba/93443 LA - eng ER - TY - JOUR TI - Learning with cone-based geometric models and orthologics AU - Leemhuis, Mena AU - Özgür, L. Özçep AU - Wolter, Diedrich PY - 2022 JO - Annals of mathematics and artificial intelligence VL - 90 IS - 11-12 SP - 1159-1195 PB - Springer Science + Business Media B.V CY - Dordrecht [u.a.] SN - 1012-2443 DO - 10.1007/s10472-022-09806-1 UR - https://fis.uni-bamberg.de/handle/uniba/91605 LA - eng AB - Recent approaches for knowledge-graph embeddings aim at connecting quantitative data structures used in machine learning to the qualitative structures of logics. Such embeddings are of a hybrid nature, they are data models that also exhibit conceptual structures inherent to logics. One motivation to investigate embeddings is to design conceptually adequate machine learning (ML) algorithms that learn or incorporate ontologies expressed in some logic. This paper investigates a new approach to embedding ontologies into geometric models that interpret concepts by geometrical structures based on convex cones. The ontologies are assumed to be represented in an orthologic, a logic with a full (ortho)negation. As a proof of concept this cone-based embedding was implemented within two ML algorithms for weak supervised multi-label learning. Both algorithms rely on cones but the first addresses ontologies expressed in classical propositional logic whereas the second addresses a weaker propositional logic, namely a weak orthologic that does not fulfil distributivity. The algorithms were evaluated and showed promising results that call for investigating other (sub)classes of cones and developing fine-tuned algorithms based on them. KW - Concept learning KW - Knowledge graph embedding KW - Support-vector machine KW - Multi-label learning KW - Orthologic ER - TY - JOUR TI - Just-In-Time Constraint-Based Inference for Qualitative Spatial and Temporal Reasoning AU - Sioutis, Michael PY - 2022 SP - 259-270 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/55184 LA - eng AB - We discuss a research roadmap for going beyond the state of the art in qualitative spatial and temporal reasoning (QSTR). Simply put, QSTR is a major field of study in Artificial Intelligence that abstracts from numerical quantities of space and time by using qualitative descriptions instead (e.g., precedes, contains, is left of); thus, it provides a concise framework that allows for rather inexpensive reasoning about entities located in space or time. Applications of QSTR can be found in a plethora of areas and domains such as smart environments, intelligent vehicles, and unmanned aircraft systems. Our discussion involves researching novel local consistencies in the aforementioned discipline, defining dynamic algorithms pertaining to these consistencies that can allow for efficient reasoning over changing spatio-temporal information, and leveraging the structures of the locally consistent related problems with regard to novel decomposability and theoretical tractability properties. Ultimately, we argue for pushing the envelope in QSTR via defining tools for tackling dynamic variants of the fundamental reasoning problems in this discipline, i.e., problems stated in terms of changing input data. Indeed, time is a continuous flow and spatial objects can change (e.g., in shape, size, or structure) as time passes; therefore, it is pertinent to be able to efficiently reason about dynamic spatio-temporal data. Finally, these tools are to be integrated into the larger context of highly active areas such as neuro-symbolic learning and reasoning, planning, data mining, and robotic applications. Our final goal is to inspire further discussion in the community about constraint-based QSTR in general, and the possible lines of future research that we outline here in particular. KW - Qualitative constraints KW - Spatio-temporal reasoning KW - Just-in-time inference KW - Local consistencies KW - Singleton checks KW - Dynamic algorithms KW - Decomposability KW - Adaptivity KW - Parallelization ER - TY - CPAPER TI - Reinforcement Learning With Imperfect Safety Constraints AU - Ro, Jin Woo AU - Lüttgen, Gerald AU - Wolter, Diedrich PY - 2022 T2 - Artificial Intelligence Safety 2022 : Proceedings of the Workshop on Artificial Intelligence Safety 2022 (SafeAI 2022) co-located with the Thirty-Sixth AAAI Conference on Artificial Intelligence (AAAI2022) Virtual, February, 2022. SP - 1-8 PB - RWTH Aachen CY - Aachen, Germany T3 - CEUR workshop proceedings UR - https://fis.uni-bamberg.de/handle/uniba/55364 LA - eng ER - TY - JOUR TI - Semi-Supervised Dimensionality Reduction by Linear Compression and Stretching AU - Long, Zhiguo AU - Meng, Hua AU - Sioutis, Michael PY - 2022 SP - 27308-27317 PB - Otto-Friedrich-Universität CY - Bamberg UR - https://fis.uni-bamberg.de/handle/uniba/53789 LA - eng AB - Dimensionality reduction is a fundamental and important research topic in the field of machine learning. This paper focuses on a dimensionality reduction technique that exploits semi-supervising information in the form of pairwise constraints; specifically, these constraints specify whether two instances belong to the same class or not. We propose two dual linear methods to accomplish dimensionality reduction under that setting. These two methods overcome the difficulty of maximizing between-class difference and minimizing within-class difference at the same time, by transforming the original data into a new space in such a way that the bi-objective problem is (almost) equivalently reduced to a single objective problem. Empirical evaluations on a broad range of public datasets show that the two proposed methods are superior to several existing methods for semi-supervised dimensionality reduction. KW - dimensionality reduction KW - principal component analysis KW - eigenvalues KW - eigenfunctions KW - task analysis KW - clustering algorithms KW - linear programming ER - TY - RPRT TI - Behind The Corner : Using Qualitative Reasoning for Solving Angry Birds AU - Haase, Felix AU - Wolter, Diedrich PY - 2021 UR - https://fis.uni-bamberg.de/handle/uniba/114328 LA - eng AB - Physics-based logic puzzle games present diverse challenges for AI, primarily related to coping with uncertainty in complex continuous environments. As these challenges tie in with the goals of qualitative reasoning (QR), in particular to plan, predict and diagnose physical mechanisms, these games pose a natural benchmark for QR. In this paper we consider the projectile collision games Angry Birds by Rovio Entertainment on which the AI bench- mark AIBirds is based. We investigate whether a qualitative approach to action planning under uncertainty proposed by Ge et al. (2016) can be adapted to the domain of Angry Birds in order to identify targets hittable by multiple rebounds. It turns out that the search space for solving Angry Birds is complexly structured and a fine-grained decomposition is required, leading to high computational costs. This paper presents an analysis of the problem and discusses means to overcome the problems faced. We propose an improved qualitative prediction method that is able to solve hard levels, yet it also suggests the need of complementary methods. KW - physical reasoning KW - qualitative simulation KW - action planning ER - TY - JOUR TI - On neighbourhood singleton-style consistencies for qualitative spatial and temporal reasoning AU - Sioutis, Michael AU - Paparrizou, Anastasia AU - Janhunen, Tomi PY - 2021 JO - Information and computation VL - 280 IS - 104638 PB - Elsevier CY - Amsterdam SN - 1090-2651 DO - 10.1016/j.ic.2020.104638 UR - https://fis.uni-bamberg.de/handle/uniba/109605 LA - eng ER - TY - CPAPER TI - Towards Robust Qualitative Spatio-Temporal Reasoning for Hybrid AI Systems AU - Sioutis, Michael AU - Meng, Hua PY - 2021 SP - 426-430 PB - IEEE DO - 10.1109/ISKE54062.2021.9755429 UR - https://fis.uni-bamberg.de/handle/uniba/56864 LA - eng ER - TY - JOUR TI - Dynamic branching in qualitative constraint-based reasoning via counting local models AU - Sioutis, Michail AU - Wolter, Diedrich PY - 2021 JO - Information and Computation VL - 281 IS - December 2021, 104787 PB - Elsevier CY - Amsterdam SN - 0890-5401 DO - 10.1016/j.ic.2021.104787 UR - https://fis.uni-bamberg.de/handle/uniba/55354 LA - eng ER - TY - CPAPER TI - On Robust Vs Fast Solving of Qualitative Constraints AU - Wehner, Jan AU - Sioutis, Michail AU - Wolter, Diedrich PY - 2021 T2 - 2021 IEEE 33rd International Conference on Tools with Artificial Intelligence : ICTAI 2021 : virtual conference, 1-3 November 2021 : proceedings SP - 801-805 PB - IEEE CY - Piscataway, NJ SN - 978-1-6654-0898-1 DO - 10.1109/ICTAI52525.2021.00128 UR - https://fis.uni-bamberg.de/handle/uniba/55353 LA - eng ER - TY - JOUR TI - A reasoning model for geo-referencing named and unnamed spatial entities in natural language place descriptions AU - Yousaf, Madiha AU - Wolter, Diedrich PY - 2021 JO - Spatial cognition and computation : an interdisciplinary journal SP - 1-39 PB - Taylor & Francis CY - London [u.a.] SN - 1542-7633 DO - 10.1080/13875868.2021.2002872 UR - https://fis.uni-bamberg.de/handle/uniba/55351 LA - eng ER - TY - CPAPER TI - Qualitative Spatial and Temporal Reasoning : Current Status and Future Challenges AU - Sioutis, Michael AU - Wolter, Diedrich PY - 2021 SP - 4594-4601 SN - 978-0-9992411-9-6 DO - 10.24963/ijcai.2021/624 UR - https://fis.uni-bamberg.de/handle/uniba/53749 LA - eng ER - TY - CHAP TI - Wissen über Raum und Zeit AU - Wolter, Diedrich AU - Özgür, L. Özçep PY - 2020 T2 - Handbuch der Künstlichen Intelligenz SP - 101-142 PB - De Gruyter CY - Berlin/Boston ET - 6 SN - 9783110659948 DO - 10.1515/9783110659948-004 UR - https://fis.uni-bamberg.de/handle/uniba/55363 LA - deu ER - TY - CPAPER TI - Dynamic Branching in Qualitative Constraint Networks via Counting Local Models AU - Sioutis, Michail AU - Wolter, Diedrich PY - 2020 T2 - 27th International Symposium on Temporal Representation and Reasoning : TIME 2020, September 23–25, 2020, Bozen-Bolzano, Italy PB - Schloss Dagstuhl - Leibniz-Zentrum für Informatik GmbH, Dagstuhl Publishing CY - Saarbrücken/Wadern T3 - LIPIcs - Leibniz International Proceedings in Informatics SN - 978-3-95977-167-2 DO - 10.4230/LIPICS.TIME.2020.12 UR - https://fis.uni-bamberg.de/handle/uniba/55358 LA - eng ER - TY - CPAPER TI - Cones, Negation, and All That AU - Lütfü Özçep, Özgür AU - Leemhuis, Mena AU - Wolter, Diedrich PY - 2020 T2 - KI 2020: Advances in Artificial Intelligence : 43rd German Conference on AI, Bamberg, Germany, September 21–25, 2020, Proceedings SP - 235-239 PB - Springer International Publishing CY - Cham T3 - Lecture Notes in Computer Science SN - 978-3-030-58285-2 DO - 10.1007/978-3-030-58285-2_17 UR - https://fis.uni-bamberg.de/handle/uniba/55357 LA - eng ER - TY - CPAPER TI - Multi-label Learning with a Cone-Based Geometric Model AU - Leemhuis, Mena AU - Lütfü Özçep, Özgür AU - Wolter, Diedrich PY - 2020 T2 - Ontologies and Concepts in Mind and Machine : 25th International Conference on Conceptual Structures, ICCS 2020, Bolzano, Italy, September 18–20, 2020, Proceedings SP - 177-185 PB - Springer CY - Cham T3 - Lecture Notes in Computer Science SN - 978-3-030-57855-8 DO - 10.1007/978-3-030-57855-8_13 UR - https://fis.uni-bamberg.de/handle/uniba/55356 LA - eng ER - TY - CPAPER TI - KI 2020: advances in artificial intelligence : 43rd German Conference on AI, Bamberg, Germany, September 21–25, 2020, proceedings ED - Schmid, Ute ED - Klügl, Franziska ED - Wolter, Diedrich PY - 2020 PB - Springer CY - Cham, Switzerland T3 - Lecture notes in computer science; Lecture notes in artificial intelligence SN - 9783030582845 DO - 10.1007/978-3-030-58285-2 UR - https://fis.uni-bamberg.de/handle/uniba/55355 LA - eng ER - TY - CPAPER TI - Cone Semantics for Logics with Negation AU - Lütfü Özçep, Özgür AU - Leemhuis, Mena AU - Wolter, Diedrich PY - 2020 T2 - Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence (IJCAI-20) SP - 1820-1826 PB - AAAI SN - 978-0-9992411-6-5 DO - 10.24963/ijcai.2020/252 UR - https://fis.uni-bamberg.de/handle/uniba/55352 LA - eng ER - TY - JOUR TI - Just-In-Time Constraint-Based Inference for Qualitative Spatial and Temporal Reasoning AU - Sioutis, Michael PY - 2020 JO - KI - Künstliche Intelligenz VL - 34 IS - 2 SP - 259-270 PB - Springer CY - Berlin SN - 1610-1987 DO - 10.1007/s13218-020-00652-z UR - https://fis.uni-bamberg.de/handle/uniba/55183 LA - eng AB - We discuss a research roadmap for going beyond the state of the art in qualitative spatial and temporal reasoning (QSTR). Simply put, QSTR is a major field of study in Artificial Intelligence that abstracts from numerical quantities of space and time by using qualitative descriptions instead (e.g., precedes, contains, is left of); thus, it provides a concise framework that allows for rather inexpensive reasoning about entities located in space or time. Applications of QSTR can be found in a plethora of areas and domains such as smart environments, intelligent vehicles, and unmanned aircraft systems. Our discussion involves researching novel local consistencies in the aforementioned discipline, defining dynamic algorithms pertaining to these consistencies that can allow for efficient reasoning over changing spatio-temporal information, and leveraging the structures of the locally consistent related problems with regard to novel decomposability and theoretical tractability properties. Ultimately, we argue for pushing the envelope in QSTR via defining tools for tackling dynamic variants of the fundamental reasoning problems in this discipline, i.e., problems stated in terms of changing input data. Indeed, time is a continuous flow and spatial objects can change (e.g., in shape, size, or structure) as time passes; therefore, it is pertinent to be able to efficiently reason about dynamic spatio-temporal data. Finally, these tools are to be integrated into the larger context of highly active areas such as neuro-symbolic learning and reasoning, planning, data mining, and robotic applications. Our final goal is to inspire further discussion in the community about constraint-based QSTR in general, and the possible lines of future research that we outline here in particular. KW - Qualitative constraints KW - Spatio-temporal reasoning KW - Just-in-time inference KW - Local consistencies KW - Singleton checks KW - Dynamic algorithms KW - Decomposability KW - Adaptivity KW - Parallelization ER - TY - JOUR TI - Semi-Supervised Dimensionality Reduction by Linear Compression and Stretching AU - Long, Zhiguo AU - Meng, Hua AU - Sioutis, Michael PY - 2020 JO - IEEE access : practical research, open solutions VL - 8 SP - 27308-27317 PB - IEEE CY - New York, NY SN - 2169-3536 DO - 10.1109/ACCESS.2020.2971562 UR - https://fis.uni-bamberg.de/handle/uniba/53412 LA - eng AB - Dimensionality reduction is a fundamental and important research topic in the field of machine learning. This paper focuses on a dimensionality reduction technique that exploits semi-supervising information in the form of pairwise constraints; specifically, these constraints specify whether two instances belong to the same class or not. We propose two dual linear methods to accomplish dimensionality reduction under that setting. These two methods overcome the difficulty of maximizing between-class difference and minimizing within-class difference at the same time, by transforming the original data into a new space in such a way that the bi-objective problem is (almost) equivalently reduced to a single objective problem. Empirical evaluations on a broad range of public datasets show that the two proposed methods are superior to several existing methods for semi-supervised dimensionality reduction. KW - dimensionality reduction KW - principal component analysis KW - eigenvalues KW - eigenfunctions KW - task analysis KW - clustering algorithms KW - linear programming ER - TY - CPAPER TI - Requirements for a model-driven cloud-native migration of monolithic web-based applications AU - Lichtenthäler, Robin AU - Prechtl, Mike AU - Schwille, Christoph AU - Schwartz, Tobias AU - Cezanne, Pascal AU - Wirtz, Guido PY - 2020 VL - 35 IS - 1-2 SP - 89-100 PB - Springer CY - Berlin ; Heidelberg SN - 2524-8510 DO - 10.1007/s00450-019-00414-9 UR - https://fis.uni-bamberg.de/handle/uniba/49185 LA - eng ER - TY - JOUR TI - Formal Representation of Qualitative Direction AU - Freksa, Christian AU - van de Ven, Jasper AU - Wolter, Diedrich PY - 2018 JO - International journal of geographical information science VL - 32 IS - 12 SP - 2514-2534 PB - Taylor & Francis CY - London SN - 1362-3087 DO - 10.1080/13658816.2017.1420794 UR - https://fis.uni-bamberg.de/handle/uniba/55359 LA - eng ER - TY - CPAPER TI - Feinstaub selbst messen: Datenqualität in einem Citizen-Science-Projekt AU - Steuer, Simon AU - Beetz, Florian AU - Geißelbrecht, Martin AU - Kaiser, F. P. AU - Nicklas, Daniela AU - Nikol, G. M. PY - 2018 T2 - Sensoren und Messsysteme : Beiträge der 19. ITG/GMA-Fachtagung 26. - 27. Juni 2018 in Nürnberg SP - 532-535 PB - VDE Verlag GmbH CY - Berlin ; Offenbach SN - 978-3-8007-4683-5 UR - https://fis.uni-bamberg.de/handle/uniba/49612 LA - deu ER - TY - CPAPER TI - A Variant of Monte-Carlo Tree Search for Referring Expression Generation AU - Schwartz, Tobias AU - Wolter, Diedrich PY - 2018 T2 - KI 2018: Advances in Artificial Intelligence : 41st German Conference on AI, Berlin, Germany, September 24–28, 2018, Proceedings SP - 315-326 PB - Springer International CY - Cham SN - 978-3-030-00110-0 DO - 10.1007/978-3-030-00111-7_27 UR - https://fis.uni-bamberg.de/handle/uniba/49549 LA - eng ER - TY - JOUR TI - Smart Environments : What is it and Why Should We Care? AU - Wolter, Diedrich AU - Kirsch, Alexandra PY - 2017 JO - Künstliche Intelligenz : KI ; Forschung, Entwicklung, Erfahrungen VL - 31 IS - 3 SP - 231-237 PB - Springer CY - Berlin SN - 1610-1987 DO - 10.1007/s13218-017-0498-4 UR - https://fis.uni-bamberg.de/handle/uniba/55360 LA - eng ER - TY - CPAPER TI - Georeferencing River Networks Using Spatial Reasoning AU - Wolter, Diedrich AU - Blank, Daniel AU - Henrich, Andreas PY - 2017 T2 - GIR'17 : Proceedings of the 11th Workshop on Geographic Information Retrieval PB - ACM CY - New York, NY, USA SN - 978-1-4503-5338-0 DO - 10.1145/3155902.3155907 UR - https://fis.uni-bamberg.de/handle/uniba/43007 LA - eng AB - This paper is concerned with automatic georeferencing of river networks from raster images such as aerial photos or maps. Determining reasonable assignments between a given network of rivers derived from a textual description and an image is subject to high combinatorial complexity and uncertainty. We investigate the application of spatial reasoning in automatic georeferencing. KW - georeferencing river networks KW - spatial reasoning KW - GIS ER - TY - JOUR TI - Probabilistic reference and grounding with PRAGR for dialogues with robots AU - Mast, Vivien AU - Falomir, Zoe AU - Wolter, Diedrich PY - 2016 JO - Journal of experimental & theoretical artificial intelligence : JETAI VL - 28 IS - 5 SP - 889-911 PB - Taylor & Francis CY - London [u.a.] SN - 1362-3079 DO - 10.1080/0952813X.2016.1154611 UR - https://fis.uni-bamberg.de/handle/uniba/55361 LA - eng ER - TY - CPAPER TI - Identifying the Geographical Scope of Prohibition Signs AU - Hopf, Konstantin AU - Dageförde, Florian AU - Wolter, Diedrich PY - 2015 T2 - Spatial Information Theory : 12th International Conference, COSIT 2015 Santa Fe, NM, USA, October 12–16, 2015, Proceedings SP - 247-267 PB - Springer International Publishing CY - Cham T3 - Lecture Notes in Computer Science SN - 978-3-319-23374-1 DO - 10.1007/978-3-319-23374-1_12 UR - https://fis.uni-bamberg.de/handle/uniba/39645 LA - eng ER - TY - CHAP TI - Algebraic Properties of Qualitative Spatio-Temporal Calculi AU - Dylla, Frank AU - Mossakowski, Till AU - Schneider, Thomas AU - Wolter, Diedrich PY - 2013 T2 - Spatial Information Theory : 11th International Conference, COSIT 2013, Scarborough, UK, September 2-6, 2013 ; proceedings SP - 516-536 PB - Springer CY - Berlin [u.a.] T3 - Lecture notes in computer science DO - 10.1007/978-3-319-01790-7_28 UR - https://fis.uni-bamberg.de/handle/uniba/20957 LA - eng AB - Qualitative spatial and temporal reasoning is based on so-called qualitative calculi. Algebraic properties of these calculi have several implications on reasoning algorithms. But what exactly is a qualitative calculus? And to which extent do the qualitative calculi proposed meet these demands? The literature provides various answers to the first question but only few facts about the second. In this paper we identify the minimal requirements to binary spatio-temporal calculi and we discuss the relevance of the according axioms for representation and reasoning. We also analyze existing qualitative calculi and provide a classification involving different notions of relation algebra. KW - Qualitative Spatial Reasoning ER -