Otevřená data na Zenodo

Tato práce aplikuje grafové neuronové sítě na predikci podcílů ve hře s názvem Ricochet Robots, NP-úplné logické hře. Herní stavy …

Tato práce aplikuje grafové neuronové sítě na predikci podcílů ve hře s názvem Ricochet Robots, NP-úplné logické hře. Herní stavy jsou reprezentovány jako orientované grafy, kde uzly odpovídají políčkům mřížky a hrany reprezentují pohyby robotů. Rekurentní architektura Graph Attention Network je trénována k napodobení hierarchické vyhledávací heuristiky, která identifikuje slibné pozice (podcíle), kterých by cílový robot měl dosáhnout. Vyhodnoceny jsou dva klasifikační úkoly: identifikace políček, ze kterých je cíl nezávisle dosažitelný, a predikce optimálních podcílů. Model dosahuje téměř dokonalého výkonu u jednoduššího úkolu dosažitelnosti a prokazuje významné učení u komplexnějšího úkolu predikce optimálních podcílů. Výsledky potvrzují, že grafové neuronové sítě dokážou zachytit prostorové uvažování potřebné pro identifikaci podcílů ve výpočetně náročných problémových doménách, čímž vytvářejí základ pro autonomní objevování podcílů v komplexních stavových prostorech. 

In this work, we pose the question of how to assess the dissimilarity of pairs of information granules that may be exemplified with I1 and I2. We focus on two representative types of information granu…

In this work, we pose the question of how to assess the dissimilarity of pairs of information granules that may be exemplified with I1 and I2. We focus on two representative types of information granules, namely fuzzy association rules (FAR) and fuzzy linguistic summaries, and aim to (1) propose a unified notation for the construction and selection of the most meaningful fuzzy information granules, and (2) analyze and discuss the assessment of dissimilarity across the considered types. 

This contribution builds upon previous achievements in the theories of generalized and intermediate quantifiers, and the evaluative linguistic expressions. In this work, we study the antonym property …

This contribution builds upon previous achievements in the theories of generalized and intermediate quantifiers, and the evaluative linguistic expressions. In this work, we study the antonym property of fuzzy linguistic summaries with absolute linguistic expressions. First, we briefly review qualitative evaluation criteria with a particular focus on the degree of truth (as baseline) and the degrees of imprecision and specificity. Next, we consider the property of antonym and investigate its adequacy for the selected criteria. 

This paper reports on the empirical performance of few-shot learning (FSL) for visualdefect classification using confidential industrial datasets. We evaluate 16 combinations offour backbone models (P…

This paper reports on the empirical performance of few-shot learning (FSL) for visualdefect classification using confidential industrial datasets. We evaluate 16 combinations offour backbone models (Perception Encoder, DINOv2, DINOv3, ConvNeXt-v2) and fourFSL classifiers (Prototypical Networks, Neighborhood Component Analysis, Relation Networks,Linear Adapter). The evaluation covers three conditions: a baseline comparison,deterministic support set augmentation, and a learnable attention preprocessor. Resultsdemonstrate that support set augmentation is a highly effective strategy, improving performancein nearly all configurations. Furthermore, the DINOv2 and ConvNeXt-V2-T backbonesemerged as top performers, achieving the most competitive and highest-accuracyresults, respectively. These findings suggest that for industrial FSL applications, combininga strong, pre-trained backbone with a simple augmentation strategy is a practicalapproach for building data-efficient classification systems. 

This paper examines the application of fuzzy natural logic in the analysis of scientific data and their representation through special linguistic expressions. We use the theory of evaluative linguisti…

This paper examines the application of fuzzy natural logic in the analysis of scientific data and their representation through special linguistic expressions. We use the theory of evaluative linguistic expressions, which makes it possible to describe quantitative data using imprecise expressions such as ``very small'', ``medium'', ``large'', and similar. They occur in the definition of the so-called intermediate quantifiers, using which we characterize given data. 

In this article, we will examine the validity of selected forms of logical syllogisms with intermediate quantifiers. We will focus in particular on forms related to the graded Peterson's cube of oppos…

In this article, we will examine the validity of selected forms of logical syllogisms with intermediate quantifiers. We will focus in particular on forms related to the graded Peterson's cube of opposition. Our verification will be based on the application of graded Peterson's rules using the distribution index. 

We thoroughly analyse the stability of sequences of FLSs with various quantifiers and various drifts, aiming to identify patterns between the statistical properties in observed multivariate time serie…

We thoroughly analyse the stability of sequences of FLSs with various quantifiers and various drifts, aiming to identify patterns between the statistical properties in observed multivariate time series and the calculated stability indexes. The secondary goal is to characterise the group of fuzzy linguistic summaries that may serve as promising explanations of changes detected in the original time series.

 In this contribution, we continue our investigation of fuzzy Peterson syllogisms. Whereas the previous study concentrated on validating these syllogisms through the construction of formal proofs…

 In this contribution, we continue our investigation of fuzzy Peterson syllogisms. Whereas the previous study concentrated on validating these syllogisms through the construction of formal proofs and semantic verification, the present work focuses on assessing their validity using Peterson’s grade-based rules. 

This paper lays the groundwork for defining division and multiplication on fuzzy intervals under complete correlation. We introduce joint possibility distributions to model dependencies between fuzzy …

This paper lays the groundwork for defining division and multiplication on fuzzy intervals under complete correlation. We introduce joint possibility distributions to model dependencies between fuzzy variables with complete correlation expressed via a linear relation. We then examine its impact on interval operations and the inverse property. Our results show that fuzzy arithmetic requires more nuanced approaches than simple point-wise interval analogies. 

In this article, we will show how relations forming the graded Peterson's square of opposition (contrary, contradictory, sub-contrary and sub-alterns) are connected with intermediate logical syllogism…

In this article, we will show how relations forming the graded Peterson's square of opposition (contrary, contradictory, sub-contrary and sub-alterns) are connected with intermediate logical syllogisms, i.e., syllogisms that contain intermediate quantifiers. Our results are related to the theory of non-monotonic reasoning.