Denna kunskapsöversikt undersöker det forskningsfält som rör musikintegrerad matematikundervisning. Syftet med översikten är att få en inblick i hur musiken påverkar elevernas matematikprestationer samt hur forskningen ser ut inom denna kombination. Därför är vår frågeställning: Vad kännetecknar forskningen om integrationen mellan matematik och musik? För att besvara denna fråga har vi utfört litteratursökningar för att finna studier och artiklar som tillsammans bildar en överblick. Med hjälp av den metod som Claes Nilholm beskriver i SMART (2016) har vi skapat en struktur för hur vi arbetat. Ur det material som vi fann under sökningarna har vi funnit mönster som talar för musikens positiva inverkan på matematikundervisning. Förmågan att uttrycka sina känslor i form av ord eller beröra andra med dem har alltid varit enbeundransvärd och sällsynt egenskap. Det här projektet handlar om att skapa en text generatorkapabel av att skriva text i stil med enastående män och kvinnor med den här egenskapen. Arbetet har genomförts genom att träna ett neuronnät med citat skrivna av märkvärdigamänniskor såsom Oscar Wilde, Mark Twain, Charles Dickens, etc. Nätverket samarbetar med två olika semantiska modeller: Word2Vec och One-Hot och alla tre är delarna som vår textgenerator består av. Med dessa genererade texterna gjordes en enkätudersökning för att samlaåsikter från studenter om kvaliteten på de genererade texterna för att på så vis utvärderalämpligheten hos de olika semantiska modellerna. Efter analysen av resultatet lärde vi oss att de flesta respondenter tyckte att texterna de läste var sammanhängande och roliga. Vi lärde oss också att Word2Vec, presterade signifikant bättre än One-hot. / The ability of expressing feelings in words or moving others with them has always been admired and rare feature. This project involves creating a text generator able to write text in the style of remarkable men and women with this ability, this gift. This has been done by training a neural network with quotes written by outstanding people such as Oscar Wilde, Mark Twain, Charles Dickens, et alt. This neural network cooperate with two different semantic models: Word2Vec and One-Hot and the three of them compound our text generator. With the text generated we carried out a survey in order to collect the opinion of students about the quality of the text generated by our generator. Upon examination of the result, we proudly learned that most of the respondents thought the texts were coherent and fun to read, we also learned that the former semantic model performed, not by a factor of magnitude, better than the latter.

Cette thèse porte sur quelques inégalités de type Trudinger-Moser et leurs applications à l'étude des injections de Sobolev qu'elles induisent dans les espaces d'Orlicz et à l'analyse d'équations aux dérivées partielles non linéaires à croissance exponentielle.Le travail qu'on présente ici se compose de trois parties. La première partie est consacrée à la description du défaut de compacité de l'injection de Sobolev 4D dans l'espace d'Orlicz dansle cadre radial.L'objectif de la deuxième partie est double. D'abord, on caractérise le défaut de compacité de l'injection de Sobolev 2D dans les différentes classes d'espaces d'Orlicz. Ensuite, on étudiel'équation de Klein-Gordon semi-linéaire avec non linéarité exponentielle, où la norme d'Orlicz joue un rôle crucial. En particulier, on aborde les questions d'existence globale, de complétude asymptotique et d'étude qualitative.Dans la troisième partie, on établit des inégalités optimales de type Adams, en étroite relation avec les inégalités de Hardy, puis on fournit une description du défaut de compacité des injections de Sobolev qu'elles induisent / This thesis focuses on some Trudinger-Moser type inequalities and their applications to the study of Sobolev embeddings they induce into the Orlicz spaces, and the investigation of nonlinear partial differential equations with exponential growth.The work presented here includes three parts. The first part is devoted to the description of the lack of compactness of the 4D Sobolev embedding into the Orlicz space in the radialframework.The aim of the second part is twofold. Firstly, we characterize the lack of compactness of the 2D Sobolev embedding into the different classes of Orlicz spaces. Secondly, we undertakethe study of the nonlinear Klein-Gordon equation with exponential growth, where the Orlicz norm plays a crucial role. In particular, issues of global existence, scattering and qualitativestudy are investigated.In the third part, we establish sharp Adams-type inequalities invoking Hardy inequalities, then we give a description of the lack of compactness of the Sobolev embeddings they induce

This thesis extends a system of an artificial player of a word-association game Codenames to easy addition of support for new languages. The system is able to play Codenames in roles as a guessing player, a clue giver or, by their combination a Duet version player. For analysis of different languages a neural toolkit Stanza was used, which is language independent and enables automated processing of many languages. It was mainly about lemmatization and part of speech tagging for selection of clues in the game. For evaluation of word associations were several models tested, where the best results had a method Pointwise Mutual Information and predictive model fastText. The system supports playing Codenames in 36 languages comprising 8 different alphabets.

The recent uprise of end-to-end speech translation models requires a new generation of parallel corpora, composed of a large amount of source language speech utterances aligned with their target language textual translations. We hereby show a pipeline and a set of methods to collect hundreds of hours of English audio-book recordings and align them with their Italian textual translations, using exclusively public domain resources gathered semi-automatically from the web. The pipeline consists in three main areas: text collection, bilingual text alignment, and forced alignment. For the text collection task, we show how to automatically find e-book titles in a target language by using machine translation, web information retrieval, and named entity recognition and translation techniques. For the bilingual text alignment task, we investigated three methods: the Gale–Church algorithm in conjunction with a small-size hand-crafted bilingual dictionary, the Gale–Church algorithm in conjunction with a bigger bilingual dictionary automatically inferred through statistical machine translation, and bilingual text alignment by computing the vector similarity of multilingual embeddings of concatenation of consecutive sentences. Our findings seem to indicate that the consecutive-sentence-embeddings similarity computation approach manages to improve the alignment of difficult sentences by indirectly performing sentence re-segmentation. For the forced alignment task, we give a theoretical overview of the preferred method depending on the properties of the text to be aligned with the audio, suggesting and using a TTS-DTW (text-to-speech and dynamic time warping) based approach in our pipeline. The result of our experiments is a publicly available multi-modal corpus composed of about 130 hours of English speech aligned with its Italian textual translation and split in 60561 triplets of English audio, English transcript, and Italian textual translation. We also post-processed the corpus so as to extract 40-MFCCs features from the audio segments and released them as a data-set.

In this master thesis we describe a method for linking named entities in a given text to a knowledge base - Named Entity Linking. Using a deep neural architecture together with BERT contextualized word embeddings we created a semi-supervised model that jointly performs Named Entity Recognition and Named Entity Disambiguation. The model outputs a Wikipedia ID for each entity detected in an input text. To compute contextualized word embeddings we used pre-trained BERT without making any changes to it (no fine-tuning). We experimented with components of our model and various versions of BERT embeddings. Moreover, we tested several different ways of using the contextual embeddings. Our model is evaluated using standard metrics and surpasses scores of models that were establishing the state of the art before the expansion of pre-trained contextualized models. The scores of our model are comparable to current state-of-the-art models.

<div> <div> <div> <div> <p>A recent surge of research on commonsense knowledge has given the AI community new opportunities and challenges. Many studies focus on constructing commonsense knowledge representations from natural language data. However, how to learn such representations from large-scale text data is still an open question. This thesis addresses the problem through statistical script learning, which learns event representations from stereotypical event relationships using weak supervision. These event representations serve as an abundant source of commonsense knowledge to be applied in downstream language tasks. We propose three script learning models that generalize previous works with new insight. A feature-enriched model characterizes fine-grained and entity-based event properties to address specific semantics. A multi-relational model generalizes traditional script learning models which rely on one type of event relationship—co-occurrence—to a multi-relational model that considers typed event relationships, going beyond simple event similarities. A narrative graph model leverages a narrative graph to inform an event with a grounded situation to maintain a global consistency of event states. Also, pretrained language models such as BERT are used to further improve event semantics.</p><p>Our three script learning models do not rely on annotated datasets, as the cost of creating these at large scales is unreasonable. Based on weak supervision, we extract events from large collections of textual data. Although noisy, the learned event representations carry profound commonsense information, enhancing performance in downstream language tasks.</p> <p>We evaluate their performance with various intrinsic and extrinsic evaluations. In the intrinsic evaluations, although the three models are evaluated in terms of various aspects, the shared core task is Multiple Choice Narrative Cloze (MCNC), which measures the model’s ability to predict what happens next, out of five candidate events, in a given situation. This task facilitates fair comparisons between script learning models for commonsense inference. The three models were proposed in three consecutive years, from 2018 to 2020, each outperforming the previous year’s model as well as the competitors’ baselines. Our best model outperforms EventComp, a widely recognized baseline, by a large margin in MCNC: i.e., absolute accuracy improvements of 9.73% (53.86% → 63.59%). In the extrinsic evaluations, we use our models for implicit discourse sense classification (IDSC), a challenging task in which two argument spans are annotated with an implicit discourse sense; the task is to predict the sense type, which requires a deep understanding of common sense between discourse arguments. Moreover, in an additional work we touch on a more interesting group of tasks about psychological commonsense reasoning. Solving these requires reasoning about and understanding human mental states such as motivation, emotion, and desire. Our best model, an enhancement of the narrative graph model, combines the advantages of the above three works to address entity-based features, typed event relationships, and grounded context in one model. The model successfully captures the context in which events appear and interactions between characters’ mental states, outperforming previous works.</p> <div> <div> <div> <p>The main contributions of this thesis are as follows: (1) We identify the importance of entity-based features for representing commonsense knowledge with script learning. (2) We create one of the first, if not the first, script learning models that addresses the multi-relational nature between events. (3) We publicly release contextualized event representations (models) trained on large-scale newswire data. (4) We develop a script learning model that combines entity-based features, typed event relationships, and grounded context in one model, and show that it is a good fit for modeling psychological common sense.</p><p>To conclude, this thesis presents an in-depth exploration of statistical script learning, enhancing existing models with new insight. Our experimental results show that models informed with the new knowledge aspects significantly outperform previous works in both intrinsic and extrinsic evaluations. </p> </div> </div> </div> </div> </div> </div> </div>

With the advancements in deep learning, object detection networks have become more robust. Nevertheless, a challenge with training deep networks is finding enough labelled training data for the model to perform well, due to constraints associated with acquiring relevant data. For this reason, active learning is used to minimize the cost by sampling the unlabeled samples that increase the performance the most. In the field of object detection, few works have been done in exploring effective hybrid active learning strategies that exploit the intermediate feature embeddings in neural networks. In this work, hybrid active learning methods are proposed and tested, using various uncertainty sampling techniques and the well-respected core-set method as the representative strategy. In addition, experiments are conducted with network embeddings to find a suitable strategy to model representation of all available samples. Experiments show mixed outcomes as to whether hybrid methods perform better than the core-set method used separately. / Med framstegen inom djupinlärning, har neurala nätverk för objektdetektering blivit mer robusta. En utmaning med att träna djupa neurala nätverk är att hitta en tillräcklig mängd träningsdata för att ett nätverk ska prestera bra, med tanke på de begränsningar som är förknippade med anskaffningen av relevant data. Av denna anledning används aktiv maskininlärning för att minimera kostnaden med att förvärva nya datapunkter, genom att göra kontinuerliga urval av de omärkta bilder som ökar prestandan mest. När det gäller objektsdetektering har få arbeten gjorts för att utforska effektiva hybridstrategier som utnyttjar de mellanliggande lagren som finns i ett neuralt nätverk. I det här arbetet föreslås och testas hybridmetoder i kontext av aktiv maskininlärning med hjälp av olika tekniker för att göra urval av datamängder baserade på osäkerhetsberäkningar men även beräkningar med hänsyn till representation (core-set-metoden). Dessutom utförs experiment med mellanliggande nätverksinbäddningar för att hitta en lämplig strategi för att modellera representation av alla tillgängliga bilder i datasetet. Experimenten visar blandade resultat när det gäller huruvida hybridmetoderna presterar bättre i jämförelse med seperata aktiv maskininlärning strategier där core-set metoden inte används.

Cross-lingual word embeddings are an increasingly important reseource in cross-lingual methods for NLP, particularly for their role in transfer learning and unsupervised machine translation, purportedly opening up the opportunity for NLP applications for low-resource languages.  However, most research in this area implicitly expects the availablility of vast monolingual corpora for training embeddings, a scenario which is not realistic for many of the world's languages.  Moreover, much of the reporting of the performance of cross-lingual word embeddings is based on a fairly narrow set of mostly European language pairs.  Our study examines the performance of cross-lingual alignment across a more diverse set of language pairs; controls for the effect of the corpus size on which the monolingual embedding spaces are trained; and studies the impact of spectral graph properties of the embedding spsace on alignment.  Through our experiments on a more diverse set of language pairs, we find that performance in bilingual lexicon induction is generally poor in heterogeneous pairs, and that even using a gold or heuristically derived dictionary has little impact on the performance on these pairs of languages.  We also find that the performance for these languages only increases slowly with corpus size.  Finally, we find a moderate correlation between the isospectral difference of the source and target embeddings and the performance of bilingual lexicon induction.  We infer that methods other than cross-lingual alignment may be more appropriate in the case of both low resource languages and heterogeneous language pairs.

Les radiologues utilisent au quotidien des solutions d'imagerie médicale pour le diagnostic. L'amélioration de l'expérience utilisateur est toujours un axe majeur de l'effort continu visant à améliorer la qualité globale et l'ergonomie des produits logiciels. Les applications de monitoring permettent en particulier d'enregistrer les actions successives effectuées par les utilisateurs dans l'interface du logiciel. Ces interactions peuvent être représentées sous forme de séquences d'actions. Sur la base de ces données, ce travail traite de deux sujets industriels : les pannes logicielles et l'ergonomie des logiciels. Ces deux thèmes impliquent d'une part la compréhension des modes d'utilisation, et d'autre part le développement d'outils de prédiction permettant soit d'anticiper les pannes, soit d'adapter dynamiquement l'interface logicielle en fonction des besoins des utilisateurs. Tout d'abord, nous visons à identifier les origines des crashes du logiciel qui sont essentielles afin de pouvoir les corriger. Pour ce faire, nous proposons d'utiliser un test binomial afin de déterminer quel type de pattern est le plus approprié pour représenter les signatures de crash. L'amélioration de l'expérience utilisateur par la personnalisation et l'adaptation des systèmes aux besoins spécifiques de l'utilisateur exige une très bonne connaissance de la façon dont les utilisateurs utilisent le logiciel. Afin de mettre en évidence les tendances d'utilisation, nous proposons de regrouper les sessions similaires. Nous comparons trois types de représentation de session dans différents algorithmes de clustering. La deuxième contribution de cette thèse concerne le suivi dynamique de l'utilisation du logiciel. Nous proposons deux méthodes -- basées sur des représentations différentes des actions d'entrée -- pour répondre à deux problématiques industrielles distinctes : la prédiction de la prochaine action et la détection du risque de crash logiciel. Les deux méthodologies tirent parti de la structure récurrente des réseaux LSTM pour capturer les dépendances entre nos données séquentielles ainsi que leur capacité à traiter potentiellement différents types de représentations d'entrée pour les mêmes données. / Radiologists use medical imaging solutions on a daily basis for diagnosis. Improving user experience is a major line of the continuous effort to enhance the global quality and usability of software products. Monitoring applications enable to record the evolution of various software and system parameters during their use and in particular the successive actions performed by the users in the software interface. These interactions may be represented as sequences of actions. Based on this data, this work deals with two industrial topics: software crashes and software usability. Both topics imply on one hand understanding the patterns of use, and on the other developing prediction tools either to anticipate crashes or to dynamically adapt software interface according to users' needs. First, we aim at identifying crash root causes. It is essential in order to fix the original defects. For this purpose, we propose to use a binomial test to determine which type of patterns is the most appropriate to represent crash signatures. The improvement of software usability through customization and adaptation of systems to each user's specific needs requires a very good knowledge of how users use the software. In order to highlight the trends of use, we propose to group similar sessions into clusters. We compare 3 session representations as inputs of different clustering algorithms. The second contribution of our thesis concerns the dynamical monitoring of software use. We propose two methods -- based on different representations of input actions -- to address two distinct industrial issues: next action prediction and software crash risk detection. Both methodologies take advantage of the recurrent structure of LSTM neural networks to capture dependencies among our sequential data as well as their capacity to potentially handle different types of input representations for the same data.

Dans le domaine du traitement automatique du langage naturelle, la tâche question-réponse (Question-Answer (QA)) consistant à développer des systèmes générant une réponse plausible à une question posée en langage naturel par un utilisateur depuis une source d’information, demeure d’actualité. Elle présente de nombreuses applications pratiques dont la recherche affinée d’information sur le web. Aujourd’hui, suite à une requête, les moteurs de recherche actuels retournent des listes classées de documents mais ils ne fournissent pas de réponse à la question de l’utilisateur. Depuis plus de cinquante ans, différentes approches et technologies ont été développées, qui ont mené à des avancées significatives en QA. Parmi celles-ci, les embeddings de mots, des vecteurs numériques représentant la signification des mots dans des contextes, des ensembles de données modèles et l’utilisation de réseaux neuronaux (RN) ont permis le développement de systèmes QA performants. Dans ce contexte, ce mémoire a porté sur l’étude du modèle BIDAF (Bidirectional Attention Flow), un des systèmes QA à base de RN les plus performants au début de ces travaux, testé au moyen de SQuAD (Stanford Question Answering Dataset), un des benchmarks les plus populaires en QA. L’étude initiale de BIDAF a démontré un certain nombre de différences structurelles notables entre sa description littéraire et son implémentation. Différentes variantes de BIDAF ont donc été développées et testées auxquelles on a ajouté l’entraînement des embeddings de mots durant l’entraînement du modèle ainsi que la modulation cyclique du taux d’apprentissage. Les modèles de structures similaires à la description littéraire de BIDAF avec entraînement des embeddings de mots et une version simplifiée ont démontré de meilleures performances, soit de 59.56% en exact match (EM) et 67.09% en F1, et EM = 60.20% et F1 = 67.69%, respectivement. Cette performance a été améliorée par la modulation du taux d’apprentissage à EM = 61.42% et F1 = 68.46%. / In the field of natural language processing, the question-answer (QA) task involving the development of systems generating a plausible answer to a question asked in natural language by a user from an information source, remains a hot topic. It has many practical applications including fine-grained information retrieval on the web. Today, following a query, current search engines return lists of classified documents but they do not provide an answer to the user’s question. Since more than fifty years, a number of approaches and technologies have been developed which have led to significant advances in QA. Among these, word embeddings, numerical vectors representing the meaning of words in contexts, the development of model or benchmark datasets and the use of neural networks (NN) have enabled the development of efficient QA systems. In this context, this thesis focused on the study of the BIDAF (Bidirectional Attention Flow) model, one of the most efficient NN-based QA systems at the start of this work, tested using SQuAD (Stanford Question Answering Dataset), one of the most popular benchmarks in QA. BIDAF’s initial study showed a number of notable structural differences between its literary description and its implementation. Different variants of BIDAF were therefore developed and tested to which were added word embedding training during model training as well as cyclic modulation of the learning rate. The models of structures similar to the literary description of BIDAF with training of word embeddings and a simplified version showed better performances, i.e. 59.56% in exact match (EM) and 67.09% in F1, and EM = 60.20% and F1 = 67.69 %, respectively. This performance was improved by modulating the learning rate to EM = 61.42% and F1 = 68.46%.