In recent years, the knowledge graph has become a commonly used storage way for large-scale knowledge in the power grid. It has proved to have an excellent performance which helps people get specialized knowledge easi...
ISBN:
(纸本)9789819947607;9789819947614
In recent years, the knowledge graph has become a commonly used storage way for large-scale knowledge in the power grid. It has proved to have an excellent performance which helps people get specialized knowledge easier. However, generating newknowledge automatically in the incomplete knowledge graph is still an urgent problem to be resolved, which name is knowledge graph completion. Previous works do not pay enough attention to the structural information in the power grid knowledge graph, resulting in poor performance. In this paper, we propose a novel framework called Complex Structure Entropy Network (CSEN) to conduct multi-hop reasoning over a power grid knowledge graph with novel two-stage cognitive theory and von Neumann graph entropy. The paper evaluates the model on the power grid defects dataset in the link prediction task and shows the effectiveness of the proposed method compared to a variety of baselines.
Cameras in sports continuously track athletes, recognize their activities, and monitor performance. This capability is ensured by sophisticated computer vision systems with machine learning algorithms and massive comp...
ISBN:
(纸本)9783031414558;9783031414565
Cameras in sports continuously track athletes, recognize their activities, and monitor performance. This capability is ensured by sophisticated computer vision systems with machine learning algorithms and massive computation power. Combat sports are rather challenging because punches happen rather quickly. This paper provides comprehensive research on approaches to measuring the performance of athletes in combat sports. We use RGB cameras to measure athletes' activity from a distance without interfering with their equipment, in contrast to the approach which uses wearable sensors. The aim of this paper is to provide a solution to classify punches in Olympic boxing based on static RGB cameras opposite the boxing ring. The proposed solution classifies three types of punches and the best classifier obtained sequentially 94%, 84%, and 81% of the F1 score for them. Finally, we measured the impact of the data augmentation process on classification performance and provide future works.
The prevalence of memes on social media has created the need to sentiment analyze their underlying meanings for censoring harmful content. Meme censoring systems by machine learning raise the need for a semi-supervise...
ISBN:
(纸本)9783031414558;9783031414565
The prevalence of memes on social media has created the need to sentiment analyze their underlying meanings for censoring harmful content. Meme censoring systems by machine learning raise the need for a semi-supervised learning solution to take advantage of the large number of unlabeled memes available on the internet and make the annotation process less challenging. Moreover, the approach needs to utilize multimodal data as memes' meanings usually come from both images and texts. This research proposes a multimodal semi-supervised learning approach that outperforms other multimodal semi-supervised learning and supervised learning state-of-the-art models on two datasets, the Multimedia Automatic Misogyny Identification and Hateful Memes dataset. Building on the insights gained from Contrastive Language-Image Pre-training, which is an effective multimodal learning technique, this research introduces SemiMemes, a novel training method that combines auto-encoder and classification task to make use of the resourceful unlabeled data.
In 2022, TIGERs Mannheim won the RoboCup Small Size League competition with individual success in the division A tournament, the blackout technical challenge and the dribbling technical challenge. The paper starts wit...
ISBN:
(纸本)9783031284687;9783031284694
In 2022, TIGERs Mannheim won the RoboCup Small Size League competition with individual success in the division A tournament, the blackout technical challenge and the dribbling technical challenge. The paper starts with an outline of the robot's dribbling hardware and ball catching computations, followed by a high level summary of the AI used in the tournament. Given 62 scored goals and no conceded goals at RoboCup 2022, the focus is on describing the used attack and support behaviors and how they are selected. The paper concludes with a statistic of the tournament backing the efficiency of our employed strategies.
We investigate the complexity of the satisfiability problem for a modal logic expressing 'knowing how' assertions, related to an agent's abilities to achieve a certain goal. We take one of the most standar...
ISBN:
(纸本)9783031436185;9783031436192
We investigate the complexity of the satisfiability problem for a modal logic expressing 'knowing how' assertions, related to an agent's abilities to achieve a certain goal. We take one of the most standard semantics for this kind of logics based on linear plans. Our main result is a proof that checking satisfiability of a 'knowing how' formula can be done in Sigma(P)(2). The algorithm we present relies on eliminating nested modalities in a formula, and then performing multiple calls to a satisfiability checking oracle for propositional logic.
Answer-Set Programming (ASP) is a popular declarative reasoning and problem solving formalism. Due to the increasing interest in explainability, several explanation approaches have been developed for ASP. However, whi...
ISBN:
(纸本)9783031436185;9783031436192
Answer-Set Programming (ASP) is a popular declarative reasoning and problem solving formalism. Due to the increasing interest in explainability, several explanation approaches have been developed for ASP. However, while those formalisms are correct and interesting on their own, most are more technical and less oriented towards philosophical or social concepts of explanation. In this work, we study the notion of contrastive explanation, i.e., answering questions of the form "Why P instead of Q?", in the context of ASP. In particular, we are interested in answering why atoms are included in an answer set, whereas others are not. Contrastive explainability has recently become popular due to its strong support from the philosophical, cognitive, and social sciences and its apparent ability to provide explanations that are concise and intuitive for humans. We formally define contrastive explanations for ASP based on counterfactual reasoning about programs. Furthermore, we demonstrate the usefulness of the concept on example applications and give some complexity results. The latter also provide a guideline as to how the explanations can be computed in practice.
In this work, we imitate the process of a legal expert studying the situational application of statutes, in order to infer relevance and entailment relationships between a query statement and a statute. While using tr...
ISBN:
(纸本)9783031291678;9783031291685
In this work, we imitate the process of a legal expert studying the situational application of statutes, in order to infer relevance and entailment relationships between a query statement and a statute. While using transformer-based architectures, we extract additional statute information from textbooks and incorporate this knowledge into the original pipeline. Results indicate that there is a benefit of using the textbook knowledge in Statute Retrieval and Entailment Classification tasks.
Xiplot is an HTML5-based system for interactive exploration of data and machine learning models. A key aspect is interaction, not only for the interactive plots but also between plots. Even though Xiplot is not restri...
ISBN:
(纸本)9783031434297;9783031434303
Xiplot is an HTML5-based system for interactive exploration of data and machine learning models. A key aspect is interaction, not only for the interactive plots but also between plots. Even though Xiplot is not restricted to any single application domain, we have developed and tested it with domain experts in quantum chemistry to study molecular interactions and regression models. Xiplot can be run both locally and online in a web browser (keeping the data local). The plots and data can also easily be exported and shared. A modular structure also makes Xiplot optimal for developing machine learning and new interaction methods.
We present ABSTRACT CNF2DDNNF, a calculus describing an approach for compiling a formula in conjunctive normal form (CNF) into deterministic negation normal form (d-DNNF). It combines component-based reasoning with a ...
ISBN:
(数字)9783031433696
ISBN:
(纸本)9783031433689;9783031433696
We present ABSTRACT CNF2DDNNF, a calculus describing an approach for compiling a formula in conjunctive normal form (CNF) into deterministic negation normal form (d-DNNF). It combines component-based reasoning with a model enumeration approach based on conflict-driven clause learning (CDCL) with chronological backtracking. Its properties, such as soundness and termination, carry over to implementations which can be modeled by it. We provide a correctness proof and a detailed example. The main conceptual differences to currently available tools targeting d-DNNF compilation are discussed and future research directions presented. The aim of this work is to lay the theoretical foundation for a novel method for d-DNNF compilation. To the best of our knowledge, our approach is the first knowledge compilation method using CDCL with chronological backtracking.
Typical online shops have one interface provided to all users, regardless of their use of the shop. Meanwhile, user behavior varies and therefore different interfaces could be provided to different user groups. Variou...
ISBN:
(纸本)9783031509582;9783031509599
Typical online shops have one interface provided to all users, regardless of their use of the shop. Meanwhile, user behavior varies and therefore different interfaces could be provided to different user groups. Various methods can be used to cluster users, including those using artificialintelligence (AI) methods. AI-based personalization allows e-commerce businesses to provide tailored recommendations to each individual customer based on preferences, purchase history, and behavior on the website. This article presents a study of the impact of an AI-based clustering method on the effectiveness of a dedicated user interface implemented and delivered to the customers of an e-shop. The first study included five methods, and two of them - agglomerative clustering and K-means clustering - were selected for detailed analysis. For both of these methods, an in-depth research was carried out and the impact of the clustering method on the quality of user clusters, as measured by the effectiveness of the dedicated interface in relation to the effectiveness of the default interface, was verified.
暂无评论