263134 research outputs found
Sort by
Probabilistische Logica om te Leren en Redeneren
No full textFrom erratic weather to unpredictable traffic jams, the world is filled with uncertainty. Reasoning with such uncertainties allows artificial intelligence (AI) systems to make decisions that are robust to unpredictable events or outcomes. This thesis focuses on methods for AI systems to handle uncertainty, in particular methods for modeling uncertainty as well as methods for reasoning and learning in such models.
More specifically, this research is situated in the field of probabilistic logic programming (PLP), a programming paradigm that combines logic and probability theory with the aim of automating inference and learning. Probability naturally expresses uncertainty, whereas logic formalizes reasoning. PLP implements these ideas with two primitives: probabilistic facts and rules. Probabilistic facts are logic statements whose truthfulness is only known with a probability. Rules allow new statements to be derived from previous statements. Because probabilistic facts are only probabilistically true, all derived statements are probabilistically true as well. The main task of PLP is to find the probability that a statement is true.
The contributions of this thesis are in three distinct domains: positive unlabeled (PU) learning, neuro-symbolic (NeSy) learning and epistemic uncertainty. The common elements over all three domains are synthesized in a general architecture for AI systems that can reason and learn with uncertainty.
PU learning is the problem of training a classifier using data with only positively labeled or unlabeled instances. Negative examples are never labeled. A method has been developped to model the uncertainty of unlabeled instances in PLP. This allows PLP methods to be used to solve PU learning problems.
NeSy systems consist of a neural and a symbolic component. The neural component usually handles perception whereas the symbolic component reasons over what is perceived. NeSy data often only contains raw input for the neural component (e.g. pixel values of an image) and a symbolic output (e.g. there are more women than men in the image), but no intermediate representation (e.g. which parts of the image depict a man). We developped a sampling based method to deal with the unknown intermediate representation when training a NeSy system.
Uncertainty can be aleatoric or epistemic in nature. Aleatoric uncertainty is fundamental uncertainty that cannot be reduced with more information. Epistemic uncertainty in contrast is uncertainty due to the lack of information. Differentiating between these two types is important to gauge the limits and confidence of AI systems. We developped new semantics that allow both aleatoric and epistemic uncertainty to be modeled in PLP. Furthemore, methods are provided for reasoning and learning with both types of uncertainty.status: Publishe
De katholieke visie op onderwijs en vorming aan de vooravond van Vaticanum II: van de pre-conciliaire desiderata tot het werk van de Commissio Praeparatoria De Studiis Et Seminariis (1959-1962)
No full textThe decree Gravissimum Educationis (GE), promulgated on October 28, 1965, deals with the issue of Catholic education and formation of the faithful, both lay and clergy. The final text does not aim to offer a detailed view on education and formation (for no agreement could be found on this matter) but offers general guidelines about education and the role of the parents, the parishes, the (non-)catholic schools, the church, and society as a whole. GE was written by the same commission that was involved in the formation of the seminarians and the document was also drafted in the context of Lumen Gentium, Apostolicam Actuositatem and Gaudium et Spes. Nonetheless, GE has not received the same attention as other documents of Vatican II. This project therefore investigates the genesis of GE which takes together the expectations of the bishops on education and formation in the pre-Vatican II period, the complex text-history of GE, the intra- and intertextuality of GE and its place within Vatican II and Catholic ecclesiology, and the reception of GE post-Vatican II. This project brings these various aspects together into a whole, and deals with each one of them in a coherent, structured and novel way.status: Publishe
Een vertaalwetenschappelijke studie van Hiëronymus' Psalmen 90-106 iuxta Hebraeos
No full textNo translation has exerted a greater influence on the history of ideas and culture than Jerome's Latin version of the Bible (ca. 347-420). Pope Benedict XV even honored Jerome with the title 'Doctor Maximus' and described him as a gift from God to the Church for the understanding of Scripture (Spiritus Paraclitus 1).
This dissertation systematically characterizes Jerome's translation technique, primarily through text-internal data collected from the translations themselves. To achieve this, it adopts and adapts an innovative framework from modern Descriptive Translation Studies—an approach that has proven fruitful in the study of other ancient Bible translations, particularly the Septuagint. A key concept in this framework is 'transformation', which this dissertation introduces for the first time in Latin Bible Studies. Transformations are non-literal translations that translators often introduce for various motivations, including linguistic constraints, stylistic preferences, coherence-related considerations, or cultural, ideological, and theological reasons. These transformations, and the motivations behind them, reveal consistent patterns and tendencies in Jerome's translation choices, suggesting the existence of translational norms within the target culture.Understanding these translational norms provides deeper insight into the shaping forces behind one of the most influential biblical translations in history.
The largest part of the dissertation consists of a detailed analysis of Jerome's translation of the fourth book of the Psalms iuxta Hebraeos, applying the developed methodology. This case study serves both as a proof of concept for the methodological framework and as the first comprehensive examination of this particular translation. The analysis generates new data that helps answer longstanding questions about Jerome's translation practice—particularly the tension between his adherence to Latin linguistic and stylistic norms and his fidelity to the Hebrew source, his command of Hebrew and reliance on Hexaplaric sources, the differences in Jerome's approach in his iuxta Hebraeos and iuxta Graecos translations of the Psalter, and the role of his theological views in his preference for particular renderings.status: Publishe
Long-term outcomes of arthroscopic Bankart repair: risk of recurrence and osteoarthritis at 11-20 years of follow-up
No full textBACKGROUND: Arthroscopic Bankart repair (ABR) is a commonly performed treatment for anterior shoulder instability. Risk factors for recurrence of dislocation after this procedure have been investigated previously at short- and mid-term follow-up. Recurrence of dislocation in the long term, as well as the incidence and risk factors for the development of osteoarthritis (OA) after this procedure require further investigation. METHODS: In this single-institution retrospective study, patients who had received ABR for structural anterior instability of the shoulder with a history of at least 1 documented dislocation, were contacted for a clinical and radiographic reassessment at long-term follow-up. Functional outcomes were assessed by the Oxford Shoulder Instability Score, Rowe score, and subjective shoulder value. Recurrence was defined as redislocation. Postoperative range of motion and strength were systematically assessed. Standard radiographs of the operated shoulder were screened for degenerative changes and scored according to the Samilson-Prieto classification for OA. RESULTS: Out of 59 consecutive shoulders eligible for inclusion, 49 shoulders were reassessed at a mean follow-up of 15 (standard deviation [SD] ± 2.4, range 11-20) years after the index procedure. The mean age of our cohort was 27.7 (SD ± 9.5) years. Nine shoulders (18.4%) sustained recurrence of dislocation. At the end of follow-up, the mean subjective shoulder value was 87% (SD ± 18, range 40-100), the mean Oxford Shoulder Instability score was 44.5 (SD ± 4.6, range 30-48) and the mean Rowe score 81.1 (SD ± 19.1, range 30-100). At final follow-up, 57% of shoulders had developed mild, 16% moderate and 11% severe degenerative changes. Age at index surgery (relative risk 2.96 [1.76-17.31] at ≥30 years) and the number of preoperative dislocations (relative risk 1.30 [1.07-1.59] if ≥5 preoperative dislocations) were significant risk factors for the development of OA. Recurrence of dislocation (P = .40) and duration of follow-up (P = .31) showed no significant association with OA. CONCLUSION: In the long term, the risk of recurrence of dislocation after ABR is low in comparison with reports on conservative treatment of anterior shoulder instability, but the incidence of OA after this procedure remains equally high. Preoperative factors significantly affect the risk of OA, rather than postoperative factors. Despite the high incidence of postoperative degenerative changes in the long term, functional outcomes remain satisfactory.sponsorship: The Institute for Orthopaedic Research and Training (IORT) reports research grant funding from the following organizations broadly related to the topic of this work: Zimmer-Biomet (EDW-ZIMMER-O2030), Johnson&Johnson (J&J Chair for Hip and Shoulder Research). None of these outside sources or funders were involved in the study design, data collection, data analysis, or the preparation or editing of the manuscript. (Zimmer-Biomet|EDW-ZIMMER-O2030, JohnsonJohnson)status: Accepte
De rol van sociale processen in de ontwikkeling van psychose
No full textWith this research, we investigate the development of psychotic-like experience in the everyday life of adolescents. We use Experience Sampling Methodology to measure mood, context, personal characteristics and the interaction between subject and context. In particular, we look at the way people cope with everyday stressful experiences.status: Publishe
Low occasion setter salience results in learning conditional stimulus partial reinforcement instead of occasion setting
No full textIn real-world settings, stimulus and outcome associations often depend on situational factors, such as Pavlovian occasion setters (OSs), which disambiguate whether a conditional stimulus (CS) will predict an outcome (unconditional stimulus; US). Whereas previous studies show that OSs are often lower in salience than CSs, no study has examined how low-salience OSs affect learning. In two conditioning experiments, we investigated this from the premise that inconsistently reinforced CSs prompt searching for additional stimuli (OSs) that indicate whether the CS will be followed by the US. Occasion setting learning was assessed using extinction rate—as partial reinforcement slows extinction relative to continuous reinforcement—and self-reported latent learning of stimuli. We hypothesized that a high-salience OS would result in faster extinction rates and occasion setting learning, whereas a low-salience OS would result in slower extinction rates and CS partial reinforcement learning. The results of Experiment 1 were mixed; there was no effect of OS salience on extinction rate, but the results for latent learning supported the hypothesis. We conducted Experiment 2 to specifically test extinction rate, and the results supported our hypothesis. The findings suggest that if a salient OS is found, occasion setting is learned; otherwise, CS partial reinforcement is learned.sponsorship: Agencia Nacional de Investigacion y Desarrollo, Grant/Award Numbers: ANID-FONDECYT 1231027, ANID-PIA/PUENTE AFB220003, ANID-SIA 85220023 (Agencia Nacional de Investigacion y Desarrollo|ANID-FONDECYT 1231027, Agencia Nacional de Investigacion y Desarrollo|ANID-PIA/PUENTE AFB220003, Agencia Nacional de Investigacion y Desarrollo|ANID-SIA 85220023)status: Publishe
Ontwikkelen van radiomicstechnieken voor contrast-enhanced mammografie In silico testen met synthetische data om zeldzame types kanker te omvatten
No full textThe difficulty of detecting and/or characterizing less common breast cancer subtypes from medical images has long been recognized. The combination of advances in x-ray imaging and artificial intelligence (AI) open up new avenues for this problem. This project will build and validate a combined deep learning and handcrafted radiomics solution for contrast enhanced mammography (CEM). The new tool will offer decision support for detection and characterization using the quantitative perfusion patterns uncovered by CEM. The large number of training images required is especially challenging for a new technique, where large clinical trials are absent. We hypothesize that data poverty can be overcome with the creation of an abundant amount of synthetic, virtual cases, including in particular difficult and less common cancer subtypes. When developing the synthetic data generation pipeline, emphasis will be laid on automation and flexibility. This will allow the use of synthetic data in virtual clinical trials and in AI development and validation. This may improve the outcomes for CEM in the radiological work-up and increase cancer information content.status: Publishe
NIEUWE GAIN-OF-FUNCTION VARIANTEN IN HET IONENKANAAL TRPM3 GEASSOCIEERD MET GLOBALE ONTWIKKELINGSSTOORNISSEN
No full textDevelopmental and epileptic encephalopathies (DEE) are a heterogenous group of severe neurodevelopmental disorders characterized by global developmental delay, moderate to severe intellectual disability, and epilepsy, variably associated with other clinical features such as continuous spike-and-wave activation during sleep (CSWS), hypotonia and facial dysmorphisms. Recent genetic studies have identified de novo heterozygous mutations in the Trpm3 gene, which encodes a Ca2+-permeable ion channel, in patients with DEE. However, the functional consequences of the mutations remained poorly understood. To investigate TRPM3 variants, we used electrophysiological and microfluorimetric techniques in a heterologous overexpression system. First, functional analyses of two disease-associated mutations (p.Val1002Met and p.Pro1102Gln) revealed a robust gain-of-function (GoF) phenotype, characterized by increased basal activity, elevated sensitivity to the endogenous neurosteroid pregnenolone sulfate (PS) and heat, and altered ligand modulation. The most recurrent mutation, p.Val1002Met, disrupted non-canonical pore gating, leading to aberrant inward cation currents and Ca2+ overload. Next, we identified seven additional heterozygous missense variants, all exhibiting a robust GoF phenotype, with increased basal activity, elevated Ca2+ influx, and enhanced responses to PS. Given the pathogenic GoF effects, we explored the therapeutic potential of the FDA-approved antiseizure medication primidone, a known TRPM3 antagonist. Primidone effectively reduced the basal activity of all mutant channels, suggesting its potential as a targeted treatment. Furthermore, we identified two children with pharmacoresistant DEE-SWAS and developmental regression, each carrying a de novo TRPM3 mutation (p.Val1002Met or p.Ser1133Pro). Both mutations resulted in a GoF phenotype. Remarkably, off-label primidone treatment halted developmental regression, improved psychomotor development, and eliminated CSWS. These findings establish TRPM3 GoF mutations as a driver of a spectrum of neurodevelopmental disorders, frequently involving cerebellar dysfunction. We propose that TRPM3 GoF mutations disrupt cerebellar and cortical development via Ca2+-mediated excitotoxicity, dysregulated neuronal proliferation, and altered network excitability. However, further studies using neuronal cultures and in vivo models are essential to validate this hypothesis and refine therapeutic strategies targeting TRPM3 dysfunction.status: Publishe
