938 resultados para Linear Attention,Conditional Language Model,Natural Language Generation,FLAX,Rare diseases


Relevância:

60.00% 60.00%

Publicador:

Resumo:

The purpose of this research was to apply model checking by using a symbolic model checker on Predicate Transition Nets (PrT Nets). A PrT Net is a formal model of information flow which allows system properties to be modeled and analyzed. The aim of this thesis was to use the modeling and analysis power of PrT nets to provide a mechanism for the system model to be verified. Symbolic Model Verifier (SMV) was the model checker chosen in this thesis, and in order to verify the PrT net model of a system, it was translated to SMV input language. A software tool was implemented which translates the PrT Net into SMV language, hence enabling the process of model checking. The system includes two parts: the PrT net editor where the representation of a system can be edited, and the translator which converts the PrT net into an SMV program.

Relevância:

60.00% 60.00%

Publicador:

Resumo:

This investigation is grounded within the concept of embodied cognition where the mind is considered to be part of a biological system. A first year undergraduate Mechanical Engineering cohort of students was tasked with explaining the behaviour of three balls of different masses being rolled down a ramp. The explanations given by the students highlighted the cognitive conflict between the everyday interpretation of the word energy and its mathematical use. The results showed that even after many years of schooling, students found it challenging to interpret the mathematics they had learned and relied upon pseudo-scientific notions to account for the behaviour of the balls.

Relevância:

60.00% 60.00%

Publicador:

Resumo:

Background: It is well documented that children with Specific Language Impairment (SLI) experience significant grammatical deficits. While much of the focus in the past has been on their morphosyntactic difficulties, less is known about their acquisition of complex syntactic structures such as relative clauses. The role of memory in language performance has also become increasingly prominent in the literature. Aims: This study aims to investigate the control of an important complex syntactic structure, the relative clause, by school age children with SLI in Ireland, using a newly devised sentence recall task. It also aims to explore the role of verbal and short-termworking memory in the performance of children with SLI on the sentence recall task, using a standardized battery of tests based on Baddeley’s model of working memory. Methods and Procedures: Thirty two children with SLI, thirty two age matched typically developing children (AM-TD) between the ages of 6 and 7,11 years and twenty younger typically developing (YTD) children between 4,7 and 5 years, completed the task. The sentence recall (SR) task included 52 complex sentences and 17 fillers. It included relative clauses that are used in natural discourse and that reflect a developmental hierarchy. The relative clauses were also controlled for length and varied in syntactic complexity, representing the full range of syntactic roles. There were seven different relative clause types attached to either the predicate nominal of a copular clause (Pn), or to the direct object of a transitive clause (Do). Responses were recorded, transcribed and entered into a database for analysis. TheWorkingMemory Test Battery for children (WMTB-C—Pickering & Gathercole, 2001) was administered in order to explore the role of short-term memory and working memory on the children’s performance on the SR task. Outcomes and Results: The children with SLI showed significantly greater difficulty than the AM-TD group and the YTD group. With the exception of the genitive subject clauses, the children with SLI scored significantly higher on all sentences containing a Pn main clause than those containing a transitive main clause. Analysis of error types revealed the frequent production of a different type of relative clause than that presented in the task—with a strong word order preference in the NVN direction indicated for the children with SLI. The SR performance for the children with SLI was most highly correlated with expressive language skills and digit recall. Conclusions and Implications: Children with SLI have significantly greater difficulty with relative clauses than YTD children who are on average two years younger—relative clauses are a delay within a delay. Unlike the YTD children they show a tendency to simplify relative clauses in the noun verb noun (NVN) direction. They show a developmental hierarchy in their production of relative clause constructions and are highly influenced by the frequency distribution of the relative clauses in the ambient language.

Relevância:

60.00% 60.00%

Publicador:

Resumo:

Provenance plays a pivotal in tracing the origin of something and determining how and why something had occurred. With the emergence of the cloud and the benefits it encompasses, there has been a rapid proliferation of services being adopted by commercial and government sectors. However, trust and security concerns for such services are on an unprecedented scale. Currently, these services expose very little internal working to their customers; this can cause accountability and compliance issues especially in the event of a fault or error, customers and providers are left to point finger at each other. Provenance-based traceability provides a mean to address part of this problem by being able to capture and query events occurred in the past to understand how and why it took place. However, due to the complexity of the cloud infrastructure, the current provenance models lack the expressibility required to describe the inner-working of a cloud service. For a complete solution, a provenance-aware policy language is also required for operators and users to define policies for compliance purpose. The current policy standards do not cater for such requirement. To address these issues, in this paper we propose a provenance (traceability) model cProv, and a provenance-aware policy language (cProvl) to capture traceability data, and express policies for validating against the model. For implementation, we have extended the XACML3.0 architecture to support provenance, and provided a translator that converts cProvl policy and request into XACML type.

Relevância:

60.00% 60.00%

Publicador:

Resumo:

International audience

Relevância:

60.00% 60.00%

Publicador:

Resumo:

In this paper we study the effect of two distinct discrete delays on the dynamics of a Wilson-Cowan neural network. This activity based model describes the dynamics of synaptically interacting excitatory and inhibitory neuronal populations. We discuss the interpretation of the delays in the language of neurobiology and show how they can contribute to the generation of network rhythms. First we focus on the use of linear stability theory to show how to destabilise a fixed point, leading to the onset of oscillatory behaviour. Next we show for the choice of a Heaviside nonlinearity for the firing rate that such emergent oscillations can be either synchronous or anti-synchronous depending on whether inhibition or excitation dominates the network architecture. To probe the behaviour of smooth (sigmoidal) nonlinear firing rates we use a mixture of numerical bifurcation analysis and direct simulations, and uncover parameter windows that support chaotic behaviour. Finally we comment on the role of delays in the generation of bursting oscillations, and discuss natural extensions of the work in this paper.

Relevância:

60.00% 60.00%

Publicador:

Resumo:

The purpose of this research was to determine if a multi-component consultation intervention was effective in improving pragmatic performance in students with ADHD. Participants for this study consisted of 7 children for whom 3 data points were obtained by a parent or 2 data points by a teacher. Changes in pragmatic performance were measured by comparing reports provided by parents or teachers pre- and post- intervention. Descriptive analysis procedures were completed to summarize changes in pragmatic behavior. Results revealed the mean overall change in pragmatic behavior for children in the MCC condition (Χ=1.133) was greater than the change seen in the CAU condition (.334) after 2 months of intervention as per parent reported data. Data indicated improvement in each behavior but incongruence between teachers and parents was found. Results support the hypothesis that the multi-component consultation intervention is effective in improving the pragmatic language performance of children with ADHD.

Relevância:

60.00% 60.00%

Publicador:

Resumo:

L'image captioning è un task di machine learning che consiste nella generazione di una didascalia, o caption, che descriva le caratteristiche di un'immagine data in input. Questo può essere applicato, ad esempio, per descrivere in dettaglio i prodotti in vendita su un sito di e-commerce, migliorando l'accessibilità del sito web e permettendo un acquisto più consapevole ai clienti con difficoltà visive. La generazione di descrizioni accurate per gli articoli di moda online è importante non solo per migliorare le esperienze di acquisto dei clienti, ma anche per aumentare le vendite online. Oltre alla necessità di presentare correttamente gli attributi degli articoli, infatti, descrivere i propri prodotti con il giusto linguaggio può contribuire a catturare l'attenzione dei clienti. In questa tesi, ci poniamo l'obiettivo di sviluppare un sistema in grado di generare una caption che descriva in modo dettagliato l'immagine di un prodotto dell'industria della moda dato in input, sia esso un capo di vestiario o un qualche tipo di accessorio. A questo proposito, negli ultimi anni molti studi hanno proposto soluzioni basate su reti convoluzionali e LSTM. In questo progetto proponiamo invece un'architettura encoder-decoder, che utilizza il modello Vision Transformer per la codifica delle immagini e GPT-2 per la generazione dei testi. Studiamo inoltre come tecniche di deep metric learning applicate in end-to-end durante l'addestramento influenzino le metriche e la qualità delle caption generate dal nostro modello.

Relevância:

60.00% 60.00%

Publicador:

Resumo:

In the last decades, Artificial Intelligence has witnessed multiple breakthroughs in deep learning. In particular, purely data-driven approaches have opened to a wide variety of successful applications due to the large availability of data. Nonetheless, the integration of prior knowledge is still required to compensate for specific issues like lack of generalization from limited data, fairness, robustness, and biases. In this thesis, we analyze the methodology of integrating knowledge into deep learning models in the field of Natural Language Processing (NLP). We start by remarking on the importance of knowledge integration. We highlight the possible shortcomings of these approaches and investigate the implications of integrating unstructured textual knowledge. We introduce Unstructured Knowledge Integration (UKI) as the process of integrating unstructured knowledge into machine learning models. We discuss UKI in the field of NLP, where knowledge is represented in a natural language format. We identify UKI as a complex process comprised of multiple sub-processes, different knowledge types, and knowledge integration properties to guarantee. We remark on the challenges of integrating unstructured textual knowledge and bridge connections with well-known research areas in NLP. We provide a unified vision of structured knowledge extraction (KE) and UKI by identifying KE as a sub-process of UKI. We investigate some challenging scenarios where structured knowledge is not a feasible prior assumption and formulate each task from the point of view of UKI. We adopt simple yet effective neural architectures and discuss the challenges of such an approach. Finally, we identify KE as a form of symbolic representation. From this perspective, we remark on the need of defining sophisticated UKI processes to verify the validity of knowledge integration. To this end, we foresee frameworks capable of combining symbolic and sub-symbolic representations for learning as a solution.

Relevância:

60.00% 60.00%

Publicador:

Resumo:

The study of ancient, undeciphered scripts presents unique challenges, that depend both on the nature of the problem and on the peculiarities of each writing system. In this thesis, I present two computational approaches that are tailored to two different tasks and writing systems. The first of these methods is aimed at the decipherment of the Linear A afraction signs, in order to discover their numerical values. This is achieved with a combination of constraint programming, ad-hoc metrics and paleographic considerations. The second main contribution of this thesis regards the creation of an unsupervised deep learning model which uses drawings of signs from ancient writing system to learn to distinguish different graphemes in the vector space. This system, which is based on techniques used in the field of computer vision, is adapted to the study of ancient writing systems by incorporating information about sequences in the model, mirroring what is often done in natural language processing. In order to develop this model, the Cypriot Greek Syllabary is used as a target, since this is a deciphered writing system. Finally, this unsupervised model is adapted to the undeciphered Cypro-Minoan and it is used to answer open questions about this script. In particular, by reconstructing multiple allographs that are not agreed upon by paleographers, it supports the idea that Cypro-Minoan is a single script and not a collection of three script like it was proposed in the literature. These results on two different tasks shows that computational methods can be applied to undeciphered scripts, despite the relatively low amount of available data, paving the way for further advancement in paleography using these methods.

Relevância:

60.00% 60.00%

Publicador:

Resumo:

Artificial Intelligence is reshaping the field of fashion industry in different ways. E-commerce retailers exploit their data through AI to enhance their search engines, make outfit suggestions and forecast the success of a specific fashion product. However, it is a challenging endeavour as the data they possess is huge, complex and multi-modal. The most common way to search for fashion products online is by matching keywords with phrases in the product's description which are often cluttered, inadequate and differ across collections and sellers. A customer may also browse an online store's taxonomy, although this is time-consuming and doesn't guarantee relevant items. With the advent of Deep Learning architectures, particularly Vision-Language models, ad-hoc solutions have been proposed to model both the product image and description to solve this problems. However, the suggested solutions do not exploit effectively the semantic or syntactic information of these modalities, and the unique qualities and relations of clothing items. In this work of thesis, a novel approach is proposed to address this issues, which aims to model and process images and text descriptions as graphs in order to exploit the relations inside and between each modality and employs specific techniques to extract syntactic and semantic information. The results obtained show promising performances on different tasks when compared to the present state-of-the-art deep learning architectures.

Relevância:

50.00% 50.00%

Publicador:

Resumo:

The objective of this study is to describe preliminary results from the cross-cultural adaptation of the Quality of Life Assessment Questionnaire, used to measure health related quality of life (HRQL) in Brazilian children aged between 5 and 11 with HIV/AIDS. The cross-cultural model evaluated the Concept, Item, Semantic and Measurement Equivalences (internal consistency and intra-observer reliability). Evaluation of the conceptual, item, semantic equivalences showed that the Portuguese version is pertinent for the Brazilian context. Four of seven domains showed internal consistency above 0.70 (α: 0.76-0.90) and five of seven revealed intra-observer reliability (ricc: 0.41-0.70). This first Portuguese version of the HRQL questionnaire can be understood as a valuable tool for assessing children's HRQL, but further studies with large samples and more robust analyses are recommended before use in the Brazilian context.