Dissertations / Theses on the topic 'DeepL Translator'

MT is becoming a powerful tool for professional translators, language service providers and common users. The present work focuses on its quality, evaluating the translations produced by two neural MT systems – i.e. Google Translate and DeepL Translator – through manual error annotation. The data set used for this task is composed of semi-specialized German texts translated into Italian. Aim of the present work is to assess the quality of MT outputs for the data set considered and obtain a detailed overview of the type of errors made by the two neural MT systems examined. The first part of this work provides a theoretical background for MT and its evaluation. Chapter 1 deals with the definition of MT and summarizes its history. Moreover, a detailed analysis of the different MT architectures is provided, as well as an overview of the possible application scenarios and the different categories of users. Chapter 2 introduces the notion of quality in the translation field and the main automatic and manual methods applied to MT quality assessment tasks. A comprehensive analysis of some of the most significant studies on neural and phrase-based MT systems output quality is then provided. The second part of this work presents a quality assessment of the output produced by two neural MT systems, i.e. Google Translation and DeepL Translator. The evaluation was performed through manual error annotation based on a fine-grained error taxonomy. Chapter 3 outlines the methodology followed during the evaluation, with a description of the dataset, the neural MT systems chosen for the study, the annotation tool and the taxonomy used during the annotation task. Chapter 4 provides the results of the evaluation and a comment thereof, offering examples extracted from the annotated data set. The final part of this work summarizes the major findings of the present contribution. Results are then discussed, with a focus on their implication for future work.

La presente tesi si pone come obiettivo quello di indagare il rapporto tra stereotipi di genere e traduzione automatica, proponendo uno studio di caso composto da frasi contenenti dei sostantivi riferiti a professioni tradotti automaticamente dall’inglese all’italiano. Il capitolo I offre una panoramica teorica sulla traduzione automatica, per poi approfondire il tema degli stereotipi di genere nella traduzione automatica. Il capitolo II si concentra invece sul rapporto tra genere e lingua, partendo dalla definizione di stereotipi e pregiudizi per passare poi alla questione dei femminili professionali. Vengono delineati gli aspetti linguistici per poi fornire una breve panoramica delle iniziative che sono state proposte nel contesto italiano e internazionale per promuovere l’uso del genere femminile nella lingua ed evitare di usarla in maniera sessista e non inclusiva. Nel capitolo III, dopo aver esposto nel dettaglio la metodologia impiegata degli studi precedenti, si presenta la struttura frasale ideata per condurre lo studio di caso. Partendo dall’elaborazione dei dati statistici sul numero di donne per ogni occupazione, si giunge alla selezione delle professioni da impiegare nello studio. Vengono inoltre illustrati gli strumenti utilizzati nell’analisi basata su corpora che verrà condotta nel capitolo IV, dove si presenta l’analisi degli output forniti da due sistemi di traduzione automatica, DeepL e Google Translate, nella combinazione linguistica inglese-italiano. L’analisi dettagliata di tutti gli aspetti della struttura frasale è corredata da tabelle e grafici esplicativi, inoltre sono presenti approfondimenti sulle traduzioni di alcune professioni di particolare rilevanza. I risultati dell’analisi verranno infine discussi in maniera approfondita, ipotizzando infine le prospettive future degli studi in questo ambito.

Il presente elaborato ha come scopo quello di analizzare il comportamento dei due traduttori automatici più utilizzati attualmente, Google Translate e DeepL, sulla base di testi pertinenti all'ambito turistico, nello specifico, le recensioni di strutture alberghiere. Dopo un breve inquadramento teorico sulla traduzione automatica e sulle principali caratteristiche dei TA sopra citati, si passerà a un esperimento di traduzione dallo spagnolo all'italiano che li coinvolgerà in maniera diretta prendendo in considerazione tre diverse tipologie di recensione. Nel secondo capitolo, verranno valutate le proposte di traduzione offerte da Google Translate e DeepL in base a problemi linguistici, testuali, extralinguistici, di intenzionalità e pragmatici sorti durante l’esperimento. A seguito di ciò, il terzo capitolo e le conclusioni finali si focalizzeranno sui limiti che compromettono maggiormente le prestazioni dei TA, prendendo come riferimento il linguaggio e le caratteristiche proprie dei testi turistici.

Sequence-to-sequence learning led to significant improvements to machine translation (MT) and automatic speech recognition (ASR) systems. These advancements were first reflected in spoken language translation (SLT) when using a cascade of (at least) ASR and MT with the new "neural" models, then by using sequence-to-sequence learning to directly translate the input audio speech into text in the target language. In this thesis we cover both approaches to the SLT task. First, we show the limits of NMT in terms of robustness to input errors when compared to the previous phrase-based state of the art. We then focus on the NMT component to achieve better translation quality with higher computational efficiency by using a network based on weakly-recurrent units. Our last work involving a cascade explores the effects on the NMT robustness when adding automatic transcripts to the training data. In order to move to the direct speech-to-text approach, we introduce MuST-C, the largest multilingual SLT corpus for training direct translation systems. MuST-C increases significantly the size of publicly available data for this task as well as their language coverage. With such availability of data, we adapted the Transformer architecture to the SLT task for its computational efficiency . Our adaptation, which we call S-Transformer, is meant to better model the audio input, and with it we set a new state of the art for MuST-C. Building on these positive results, we finally use S-Transformer with different data applications: i) one-to-many multilingual translation by training it on MuST-C; ii participation to the IWSLT 19 shared task with data augmentation; and iii) instance-based adaptation for using the training data at test time. The results in this thesis show a steady quality improvement in direct SLT. Our hope is that the presented resources and technological solutions will increase its adoption in the near future, so to make multilingual information access easier in a globalized world.

With the advent of Deep Learning, we have been able to find solutions to many problems which didn't have an algorithmic solution, among these the image-to-image translation problem. One approach to solve it is by using the CycleGAN framework, which allows to learn mappings between two given image categories (or classes) that aren't necessarily paired. In this dissertation we present some attempts that were done in order to use the CycleGAN approach to perform image translations between more than two image classes at a time.

Automatic Post-Editing (APE) aims to correct systematic errors in a machine translated text. This is primarily useful when the machine translation (MT) system is not accessible for improvement, leaving APE as a viable option to improve translation quality as a downstream task - which is the focus of this thesis. This field has received less attention compared to MT due to several reasons, which include: the limited availability of data to perform a sound research, contrasting views reported by different researchers about the effectiveness of APE, and limited attention from the industry to use APE in current production pipelines. In this thesis, we perform a thorough investigation of APE as a down- stream task in order to: i) understand its potential to improve translation quality; ii) advance the core technology - starting from classical methods to recent deep-learning based solutions; iii) cope with limited and sparse data; iv) better leverage multiple input sources; v) mitigate the task-specific problem of over-correction; vi) enhance neural decoding to leverage external knowledge; and vii) establish an online learning framework to handle data diversity in real-time. All the above contributions are discussed across several chapters, and most of them are evaluated in the APE shared task organized each year at the Conference on Machine Translation. Our efforts in improving the technology resulted in the best system at the 2017 APE shared task, and our work on online learning received a distinguished paper award at the Italian Conference on Computational Linguistics. Overall, outcomes and findings of our work have boost interest among researchers and attracted industries to examine this technology to solve real-word problems.

La traduction automatique vise à traduire des documents d’une langue à une autre sans l’intervention humaine. Avec l’apparition des réseaux de neurones profonds (DNN), la traduction automatique neuronale(NMT) a commencé à dominer le domaine, atteignant l’état de l’art pour de nombreuses langues. NMT a également ravivé l’intérêt pour la traduction basée sur l’interlangue grâce à la manière dont elle place la tâche dans un cadre encodeur-décodeur en passant par des représentations latentes. Combiné avec la flexibilité architecturale des DNN, ce cadre a aussi ouvert une piste de recherche sur la multimodalité, ayant pour but d’enrichir les représentations latentes avec d’autres modalités telles que la vision ou la parole, par exemple. Cette thèse se concentre sur la traduction automatique multimodale(MMT) en intégrant la vision comme une modalité secondaire afin d’obtenir une meilleure compréhension du langage, ancrée de façon visuelle. J’ai travaillé spécifiquement avec un ensemble de données contenant des images et leurs descriptions traduites, où le contexte visuel peut être utile pour désambiguïser le sens des mots polysémiques, imputer des mots manquants ou déterminer le genre lors de la traduction vers une langue ayant du genre grammatical comme avec l’anglais vers le français. Je propose deux approches principales pour intégrer la modalité visuelle : (i) un mécanisme d’attention multimodal qui apprend à prendre en compte les représentations latentes des phrases sources ainsi que les caractéristiques visuelles convolutives, (ii) une méthode qui utilise des caractéristiques visuelles globales pour amorcer les encodeurs et les décodeurs récurrents. Grâce à une évaluation automatique et humaine réalisée sur plusieurs paires de langues, les approches proposées se sont montrées bénéfiques. Enfin,je montre qu’en supprimant certaines informations linguistiques à travers la dégradation systématique des phrases sources, la véritable force des deux méthodes émerge en imputant avec succès les noms et les couleurs manquants. Elles peuvent même traduire lorsque des morceaux de phrases sources sont entièrement supprimés
Machine translation aims at automatically translating documents from one language to another without human intervention. With the advent of deep neural networks (DNN), neural approaches to machine translation started to dominate the field, reaching state-ofthe-art performance in many languages. Neural machine translation (NMT) also revived the interest in interlingual machine translation due to how it naturally fits the task into an encoder-decoder framework which produces a translation by decoding a latent source representation. Combined with the architectural flexibility of DNNs, this framework paved the way for further research in multimodality with the objective of augmenting the latent representations with other modalities such as vision or speech, for example. This thesis focuses on a multimodal machine translation (MMT) framework that integrates a secondary visual modality to achieve better and visually grounded language understanding. I specifically worked with a dataset containing images and their translated descriptions, where visual context can be useful forword sense disambiguation, missing word imputation, or gender marking when translating from a language with gender-neutral nouns to one with grammatical gender system as is the case with English to French. I propose two main approaches to integrate the visual modality: (i) a multimodal attention mechanism that learns to take into account both sentence and convolutional visual representations, (ii) a method that uses global visual feature vectors to prime the sentence encoders and the decoders. Through automatic and human evaluation conducted on multiple language pairs, the proposed approaches were demonstrated to be beneficial. Finally, I further show that by systematically removing certain linguistic information from the input sentences, the true strength of both methods emerges as they successfully impute missing nouns, colors and can even translate when parts of the source sentences are completely removed

Drug development is a protracted and expensive process. One of the main challenges indrug discovery is to find molecules with desirable properties. Molecular optimization is thetask of optimizing precursor molecules by affording them with desirable properties. Recentadvancement in Artificial Intelligence, has led to deep learning models designed for molecularoptimization. These models, that generates new molecules with desirable properties, have thepotential to accelerate the drug discovery. In this thesis, I evaluate the current state-of-the-art graph-to-graph translation model formolecular optimization, the HierG2G. I examine the HierG2G’s performance using three testcases, where the second test is designed, with the help of chemical experts, to represent a commonmolecular optimization task. The third test case, tests the HierG2G’s performance on,for the model, previously unseen molecules. I conclude that, in each of the test cases, the HierG2Gcan successfully generate structurally similar molecules with desirable properties givena source molecule and an user-specified desired property change. Further, I benchmark the HierG2Gagainst two famous string-based models, the seq2seq and the Transformer. My resultsuggests that the seq2seq is the overall best model for molecular optimization, but due to thevarying performance among the models, I encourage a potential user to simultaneously use allthree models for molecular optimization.

La diversité des langues complexifie la tâche de communication entre les humains à travers les différentes cultures. La traduction automatique est un moyen rapide et peu coûteux pour simplifier la communication interculturelle. Récemment, laTraduction Automatique Neuronale (NMT) a atteint des résultats impressionnants. Cette thèse s'intéresse à la Traduction Automatique Neuronale Factorisé (FNMT) qui repose sur l'idée d'utiliser la morphologie et la décomposition grammaticale des mots (lemmes et facteurs linguistiques) dans la langue cible. Cette architecture aborde deux défis bien connus auxquelles les systèmes NMT font face. Premièrement, la limitation de la taille du vocabulaire cible, conséquence de la fonction softmax, qui nécessite un calcul coûteux à la couche de sortie du réseau neuronale, conduisant à un taux élevé de mots inconnus. Deuxièmement, le manque de données adéquates lorsque nous sommes confrontés à un domaine spécifique ou une langue morphologiquement riche. Avec l'architecture FNMT, toutes les inflexions des mots sont prises en compte et un vocabulaire plus grand est modélisé tout en gardant un coût de calcul similaire. De plus, de nouveaux mots non rencontrés dans les données d'entraînement peuvent être générés. Dans ce travail, j'ai développé différentes architectures FNMT en utilisant diverses dépendances entre les lemmes et les facteurs. En outre, j'ai amélioré la représentation de la langue source avec des facteurs. Le modèle FNMT est évalué sur différentes langues dont les plus riches morphologiquement. Les modèles à l'état de l'art, dont certains utilisant le Byte Pair Encoding (BPE) sont comparés avec le modèle FNMT en utilisant des données d'entraînement de petite et de grande taille. Nous avons constaté que les modèles utilisant les facteurs sont plus robustes aux conditions d'entraînement avec des faibles ressources. Le FNMT a été combiné avec des unités BPE permettant une amélioration par rapport au modèle FNMT entrainer avec des données volumineuses. Nous avons expérimenté avec dfférents domaines et nous avons montré des améliorations en utilisant les modèles FNMT. De plus, la justesse de la morphologie est mesurée à l'aide d'un ensemble de tests spéciaux montrant l'avantage de modéliser explicitement la morphologie de la cible. Notre travail montre les bienfaits de l'applicationde facteurs linguistiques dans le NMT
Communication between humans across the lands is difficult due to the diversity of languages. Machine translation is a quick and cheap way to make translation accessible to everyone. Recently, Neural Machine Translation (NMT) has achievedimpressive results. This thesis is focus on the Factored Neural Machine Translation (FNMT) approach which is founded on the idea of using the morphological and grammatical decomposition of the words (lemmas and linguistic factors) in the target language. This architecture addresses two well-known challenges occurring in NMT. Firstly, the limitation on the target vocabulary size which is a consequence of the computationally expensive softmax function at the output layer of the network, leading to a high rate of unknown words. Secondly, data sparsity which is arising when we face a specific domain or a morphologically rich language. With FNMT, all the inflections of the words are supported and larger vocabulary is modelled with similar computational cost. Moreover, new words not included in the training dataset can be generated. In this work, I developed different FNMT architectures using various dependencies between lemmas and factors. In addition, I enhanced the source language side also with factors. The FNMT model is evaluated on various languages including morphologically rich ones. State of the art models, some using Byte Pair Encoding (BPE) are compared to the FNMT model using small and big training datasets. We found out that factored models are more robust in low resource conditions. FNMT has been combined with BPE units performing better than pure FNMT model when trained with big data. We experimented with different domains obtaining improvements with the FNMT models. Furthermore, the morphology of the translations is measured using a special test suite showing the importance of explicitly modeling the target morphology. Our work shows the benefits of applying linguistic factors in NMT

In this work, we propose a novel and effective method for constraining the output space of the ill-posed problem of unsupervised image-to-image translation. We make the assumption that the environment of the source domain is known, and we propose to explicitly enforce preservation of the ground-truth labels on the images translated from the source to the target domain. We run empirical experiments on preserving information such as semantic segmentation and disparity and show evidence that our method achieves improved performance over the baseline model UNIT on translating images from SYNTHIA to Cityscapes. The generated images are perceived as more realistic in human surveys and have reduced errors when using them as adapted images in the domain adaptation scenario. Moreover, the underlying ground-truth preservation assumption is complementary to alternative approaches and by combining it with the UNIT framework, we improve the results even further.
I det här arbetet föreslår vi en ny och effektiv metod för att begränsa värdemängden för det illa-definierade problemet som utgörs av oövervakad bild-till-bild-översättning. Vi antar att miljön i källdomänen är känd, och vi föreslår att uttryckligen framtvinga bevarandet av grundfaktaetiketterna på bilder översatta från källa till måldomän. Vi utför empiriska experiment där information som semantisk segmentering och skillnad bevaras och visar belägg för att vår metod uppnår förbättrad prestanda över baslinjemetoden UNIT på att översätta bilder från SYNTHIA till Cityscapes. De genererade bilderna uppfattas som mer realistiska i undersökningar där människor tillfrågats och har minskat fel när de används som anpassade bilder i domänpassningsscenario. Dessutom är det underliggande grundfaktabevarande antagandet kompletterat med alternativa tillvägagångssätt och genom att kombinera det med UNIT-ramverket förbättrar vi resultaten ytterligare.

Thesis: Ph. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2018.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 183-228).
Language technology has become pervasive in everyday life. Neural networks are a key component in this technology thanks to their ability to model large amounts of data. Contrary to traditional systems, models based on deep neural networks (a.k.a. deep learning) can be trained in an end-to-end fashion on input-output pairs, such as a sentence in one language and its translation in another language, or a speech utterance and its transcription. The end-to-end training paradigm simplifies the engineering process while giving the model flexibility to optimize for the desired task. This, however, often comes at the expense of model interpretability: understanding the role of different parts of the deep neural network is difficult, and such models are sometimes perceived as "black-box", hindering research efforts and limiting their utility to society. This thesis investigates what kind of linguistic information is represented in deep learning models for written and spoken language. In order to study this question, I develop a unified methodology for evaluating internal representations in neural networks, consisting of three steps: training a model on a complex end-to-end task; generating feature representations from different parts of the trained model; and training classifiers on simple supervised learning tasks using the representations. I demonstrate the approach on two core tasks in human language technology: machine translation and speech recognition. I perform a battery of experiments comparing different layers, modules, and architectures in end-to-end models that are trained on these tasks, and evaluate their quality at different linguistic levels. First, I study how neural machine translation models learn morphological information. Second, I compare lexical semantic and part-of-speech information in neural machine translation. Third, I investigate where syntactic and semantic structures are captured in these models. Finally, I explore how end-to-end automatic speech recognition models encode phonetic information. The analyses illuminate the inner workings of end-to-end machine translation and speech recognition systems, explain how they capture different language properties, and suggest potential directions for improving them. I also point to open questions concerning the representation of other linguistic properties, the investigation of different models, and the use of other analysis methods. Taken together, this thesis provides a comprehensive analysis of internal language representations in deep learning models.
by Yonatan Belinkov.
Ph. D.

ABSTRACT The modern world is becoming every day more and more tied to technology. The technological development of the last thirty years has led us to live in a digital world dominated by the Internet, which today reaches more than 4 billion users worldwide. Thus computers, smartphones and tablets are now part of everyday life for more and more people. Behind the worldwide spread of modern technologies we find localization, which plays a key role in our society, despite being barely known by non-experts. Every day localization enables users all around the world to access digital products and content in their own language, thus becoming essential in today’s digital world. Given the author’s interest in translation technologies, the subject of this thesis is the localization of the guide memoQ 8.7 Getting Started from English into Italian. The guide is part of the user documentation of memoQ, a software for Computer-Assisted Translation (CAT) used by many professional translators. Chapter 1 starts by introducing the historical background of localization, from its origin to the consolidation of the localization industry, and then provides an overview of previous literature, focusing mainly on the process of software localization. Chapter 2 presents the localization project and all the activities carried out prior to the translation, including the analysis of the source text. The localization of the memoQ guide is the focus of Chapter 3, in which some of the most interesting aspects that emerged in the translation process are examined. Finally, Chapter 4 addresses the increasingly popular topic of Machine Translation (MT): after providing a general overview, the human translation of the memoQ guide is compared to the translation performed by the MT system DeepL.

This thesis proposes a deep learning approach to predict stress at various stages of mechanical loading in 2-D representations of fiber-reinforced composites. More specifically, the full-field stress distribution at elastic and at an early stage of damage initiation is predicted based on the microstructural geometry. The required data set for the purposes of training and validation are generated via high-fidelity simulations of several randomly generated microstructural representations with complex geometries. Two deep learning approaches are employed and their performances are compared: fully convolutional generator and Pix2Pix translation. It is shown that both the utilized approaches can well predict the stress distributions at the designated loading stages with high accuracy.
M.S.
Fiber-reinforced composites are material types with excellent mechanical performance. They form the major material in the construction of space shuttles, aircraft, fancy cars, etc., the structures that are designed to be lightweight and at the same time extremely stiff and strong. Due to the broad application, especially in the sensitives industries, fiber-reinforced composites have always been a subject of meticulous research studies. The research studies to better understand the mechanical behavior of these composites has to be conducted on the micro-scale. Since the experimental studies on micro-scale are expensive and extremely limited, numerical simulations are normally adopted. Numerical simulations, however, are complex, time-consuming, and highly computationally expensive even when run on powerful supercomputers. Hence, this research aims to leverage artificial intelligence to reduce the complexity and computational cost associated with the existing high-fidelity simulation techniques. We propose a robust deep learning framework that can be used as a replacement for the conventional numerical simulations to predict important mechanical attributes of the fiber-reinforced composite materials on the micro-scale. The proposed framework is shown to have high accuracy in predicting complex phenomena including stress distributions at various stages of mechanical loading.

This thesis introduces the concept of an encoder-decoder neural network and develops architectures for the construction of such networks. Encoder-decoder neural networks are probabilistic conditional generative models of high-dimensional structured items such as natural language utterances and natural images. Encoder-decoder neural networks estimate a probability distribution over structured items belonging to a target set conditioned on structured items belonging to a source set. The distribution over structured items is factorized into a product of tractable conditional distributions over individual elements that compose the items. The networks estimate these conditional factors explicitly. We develop encoder-decoder neural networks for core tasks in natural language processing and natural image and video modelling. In Part I, we tackle the problem of sentence modelling and develop deep convolutional encoders to classify sentences; we extend these encoders to models of discourse. In Part II, we go beyond encoders to study the longstanding problem of translating from one human language to another. We lay the foundations of neural machine translation, a novel approach that views the entire translation process as a single encoder-decoder neural network. We propose a beam search procedure to search over the outputs of the decoder to produce a likely translation in the target language. Besides known recurrent decoders, we also propose a decoder architecture based solely on convolutional layers. Since the publication of these new foundations for machine translation in 2013, encoder-decoder translation models have been richly developed and have displaced traditional translation systems both in academic research and in large-scale industrial deployment. In services such as Google Translate these models process in the order of a billion translation queries a day. In Part III, we shift from the linguistic domain to the visual one to study distributions over natural images and videos. We describe two- and three- dimensional recurrent and convolutional decoder architectures and address the longstanding problem of learning a tractable distribution over high-dimensional natural images and videos, where the likely samples from the distribution are visually coherent. The empirical validation of encoder-decoder neural networks as state-of- the-art models of tasks ranging from machine translation to video prediction has a two-fold significance. On the one hand, it validates the notions of assigning probabilities to sentences or images and of learning a distribution over a natural language or a domain of natural images; it shows that a probabilistic principle of compositionality, whereby a high- dimensional item is composed from individual elements at the encoder side and whereby a corresponding item is decomposed into conditional factors over individual elements at the decoder side, is a general method for modelling cognition involving high-dimensional items; and it suggests that the relations between the elements are best learnt in an end-to-end fashion as non-linear functions in distributed space. On the other hand, the empirical success of the networks on the tasks characterizes the underlying cognitive processes themselves: a cognitive process as complex as translating from one language to another that takes a human a few seconds to perform correctly can be accurately modelled via a learnt non-linear deterministic function of distributed vectors in high-dimensional space.

We expand the scope of image-to-image translation to include more distinct image domains, where the image sets have analogous structures, but may not share object types between them. Semantic-Driven Unsupervised Image-to-Image Translation for Distinct Image Domains (SUNIT) is built to more successfully translate images in this setting, where content from one domain is not found in the other. Our method trains an image translation model by learning encodings for semantic segmentations of images. These segmentations are translated between image domains to learn meaningful mappings between the structures in the two domains. The translated segmentations are then used as the basis for image generation. Beginning image generation with encoded segmentation information helps maintain the original structure of the image. We qualitatively and quantitatively show that SUNIT improves image translation outcomes, especially for image translation tasks where the image domains are very distinct.

Training data is an essential ingredient within supervised learning, yet time con-suming, expensive and for some applications impossible to retrieve. Thus it isof interest to use synthetic training data. However, the domain shift of syntheticdata makes it challenging to obtain good results when used as training data fordeep learning models. It is therefore of interest to refine synthetic data, e.g. using image-to-image translation, to improve results. The aim of this work is to compare different methods to do image-to-image translation of synthetic training data of thermal IR-images using GANs. Translation is done both using synthetic thermal IR-images alone, as well as including pixelwise depth and/or semantic information. To evaluate, a new measure based on the Frechét Inception Distance, adapted to work for thermal IR-images is proposed. The results show that the model trained using IR-images alone translates the generated images closest to the domain of authentic thermal IR-images. The training where IR-images are complemented by corresponding pixelwise depth data performs second best. However, given more training time, inclusion of depth data has the potential to outperform training withirdata alone. This gives a valuable insight on how to best translate images from the domain of synthetic IR-images to that of authentic IR-images, which is vital for quick and low cost generation of training data for deep learning models.

Generative Adversarial Networks (GANs) is a deep learning method that has been developed for synthesizing data. One application for which it can be used for is image-to-image translations. This could prove to be valuable when training deep neural networks for image classification tasks. Two areas where deep learning methods are used are automotive vision systems and medical imaging. Automotive vision systems are expected to handle a broad range of scenarios which demand training data with a high diversity. The scenarios in the medical field are fewer but the problem is instead that it is difficult, time consuming and expensive to collect training data. This thesis evaluates different GAN models by comparing synthetic MR images produced by the models against ground truth images. A perceptual study is also performed by an expert in the field. It is shown by the study that the implemented GAN models can synthesize visually realistic MR images. It is also shown that models producing more visually realistic synthetic images not necessarily have better results in quantitative error measurements, when compared to ground truth data. Along with the investigations on medical images, the thesis explores the possibilities of generating synthetic street view images of different resolution, light and weather conditions. Different GAN models have been compared, implemented with our own adjustments, and evaluated. The results show that it is possible to create visually realistic images for different translations and image resolutions.

This thesis is a continuation of my bachelor thesis, which is dedicated to syntax analysis based on deep pushdown automata. In theorical part of this thesis is defined everything fundamental for this work, for example deep syntax-directed translation, pushdown automata, deep pushdown automata, finite transducer and deep pushdown transducer.   The second part of this thesis is dedicated to the educational program for students of IFJ. In this part is defined strucure of this program and its parts. All part of program are analyzed from a theoretical and practical point of view.

[ES] El problema conocido como de secuencia a secuencia consiste en transformar una secuencia de entrada en una secuencia de salida. Bajo esta perspectiva se puede atacar una amplia cantidad de problemas, entre los cuales destacan la traducción automática o la descripción automática de objetos multimedia. La aplicación de redes neuronales profundas ha revolucionado esta disciplina, y se han logrado avances notables. Pero los sistemas automáticos todavía producen predicciones que distan mucho de ser perfectas. Para obtener predicciones de gran calidad, los sistemas automáticos se utilizan bajo la supervisión de un humano, quien corrige los errores. Esta tesis se centra principalmente en el problema de la traducción del lenguaje natural, usando modelos enteramente neuronales. Nuestro objetivo es desarrollar sistemas de traducción neuronal más eficientes. asentándonos sobre dos pilares fundamentales: cómo utilizar el sistema de una forma más eficiente y cómo aprovechar datos generados durante la fase de explotación del mismo. En el primer caso, aplicamos el marco teórico conocido como predicción interactiva a la traducción automática neuronal. Este proceso consiste en integrar usuario y sistema en un proceso de corrección cooperativo, con el objetivo de reducir el esfuerzo humano empleado en obtener traducciones de alta calidad. Desarrollamos distintos protocolos de interacción para dicha tecnología, aplicando interacción basada en prefijos y en segmentos, implementados modificando el proceso de búsqueda del sistema. Además, ideamos mecanismos para obtener una interacción con el sistema más precisa, manteniendo la velocidad de generación del mismo. Llevamos a cabo una extensa experimentación, que muestra el potencial de estas técnicas: superamos el estado del arte anterior por un gran margen y observamos que nuestros sistemas reaccionan mejor a las interacciones humanas. A continuación, estudiamos cómo mejorar un sistema neuronal mediante los datos generados como subproducto de este proceso de corrección. Para ello, nos basamos en dos paradigmas del aprendizaje automático: el aprendizaje muestra a muestra y el aprendizaje activo. En el primer caso, el sistema se actualiza inmediatamente después de que el usuario corrige una frase, aprendiendo de una manera continua a partir de correcciones, evitando cometer errores previos y especializándose en un usuario o dominio concretos. Evaluamos estos sistemas en una gran cantidad de situaciones y dominios diferentes, que demuestran el potencial que tienen los sistemas adaptativos. También llevamos a cabo una evaluación humana, con traductores profesionales. Éstos quedaron muy satisfechos con el sistema adaptativo. Además, fueron más eficientes cuando lo usaron, comparados con un sistema estático. El segundo paradigma lo aplicamos en un escenario en el que se deban traducir grandes cantidades de frases, siendo inviable la supervisión de todas. El sistema selecciona aquellas muestras que vale la pena supervisar, traduciendo el resto automáticamente. Aplicando este protocolo, redujimos de aproximadamente un cuarto el esfuerzo humano necesario para llegar a cierta calidad de traducción. Finalmente, atacamos el complejo problema de la descripción de objetos multimedia. Este problema consiste en describir en lenguaje natural un objeto visual, una imagen o un vídeo. Comenzamos con la tarea de descripción de vídeos pertenecientes a un dominio general. A continuación, nos movemos a un caso más específico: la descripción de eventos a partir de imágenes egocéntricas, capturadas a lo largo de un día. Buscamos extraer relaciones entre eventos para generar descripciones más informadas, desarrollando un sistema capaz de analizar un mayor contexto. El modelo con contexto extendido genera descripciones de mayor calidad que un modelo básico. Por último, aplicamos la predicción interactiva a estas tareas multimedia, disminuyendo el esfuerzo necesa
[CAT] El problema conegut com a de seqüència a seqüència consisteix en transformar una seqüència d'entrada en una seqüència d'eixida. Seguint aquesta perspectiva, es pot atacar una àmplia quantitat de problemes, entre els quals destaquen la traducció automàtica, el reconeixement automàtic de la parla o la descripció automàtica d'objectes multimèdia. L'aplicació de xarxes neuronals profundes ha revolucionat aquesta disciplina, i s'han aconseguit progressos notables. Però els sistemes automàtics encara produeixen prediccions que disten molt de ser perfectes. Per a obtindre prediccions de gran qualitat, els sistemes automàtics són utilitzats amb la supervisió d'un humà, qui corregeix els errors. Aquesta tesi se centra principalment en el problema de la traducció de llenguatge natural, el qual s'ataca emprant models enterament neuronals. El nostre objectiu principal és desenvolupar sistemes més eficients. Per a aquesta tasca, les nostres contribucions s'assenten sobre dos pilars fonamentals: com utilitzar el sistema d'una manera més eficient i com aprofitar dades generades durant la fase d'explotació d'aquest. En el primer cas, apliquem el marc teòric conegut com a predicció interactiva a la traducció automàtica neuronal. Aquest procés consisteix en integrar usuari i sistema en un procés de correcció cooperatiu, amb l'objectiu de reduir l'esforç humà emprat per obtindre traduccions d'alta qualitat. Desenvolupem diferents protocols d'interacció per a aquesta tecnologia, aplicant interacció basada en prefixos i en segments, implementats modificant el procés de cerca del sistema. A més a més, busquem mecanismes per a obtindre una interacció amb el sistema més precisa, mantenint la velocitat de generació. Duem a terme una extensa experimentació, que mostra el potencial d'aquestes tècniques: superem l'estat de l'art anterior per un gran marge i observem que els nostres sistemes reaccionen millor a les interacciones humanes. A continuació, estudiem com millorar un sistema neuronal mitjançant les dades generades com a subproducte d'aquest procés de correcció. Per a això, ens basem en dos paradigmes de l'aprenentatge automàtic: l'aprenentatge mostra a mostra i l'aprenentatge actiu. En el primer cas, el sistema s'actualitza immediatament després que l'usuari corregeix una frase. Per tant, el sistema aprén d'una manera contínua a partir de correccions, evitant cometre errors previs i especialitzant-se en un usuari o domini concrets. Avaluem aquests sistemes en una gran quantitat de situacions i per a dominis diferents, que demostren el potencial que tenen els sistemes adaptatius. També duem a terme una avaluació amb traductors professionals, qui varen quedar molt satisfets amb el sistema adaptatiu. A més, van ser més eficients quan ho van usar, si ho comparem amb el sistema estàtic. Pel que fa al segon paradigma, l'apliquem per a l'escenari en el qual han de traduir-se grans quantitats de frases, i la supervisió de totes elles és inviable. En aquest cas, el sistema selecciona les mostres que paga la pena supervisar, traduint la resta automàticament. Aplicant aquest protocol, reduírem en aproximadament un quart l'esforç necessari per a arribar a certa qualitat de traducció. Finalment, ataquem el complex problema de la descripció d'objectes multimèdia. Aquest problema consisteix en descriure, en llenguatge natural, un objecte visual, una imatge o un vídeo. Comencem amb la tasca de descripció de vídeos d'un domini general. A continuació, ens movem a un cas més específic: la descripció d''esdeveniments a partir d'imatges egocèntriques, capturades al llarg d'un dia. Busquem extraure relacions entre ells per a generar descripcions més informades, desenvolupant un sistema capaç d'analitzar un major context. El model amb context estés genera descripcions de major qualitat que el model bàsic. Finalment, apliquem la predicció interactiva a aquestes tasques multimèdia, di
[EN] The sequence-to-sequence problem consists in transforming an input sequence into an output sequence. A variety of problems can be posed in these terms, including machine translation, speech recognition or multimedia captioning. In the last years, the application of deep neural networks has revolutionized these fields, achieving impressive advances. However and despite the improvements, the output of the automatic systems is still far to be perfect. For achieving high-quality predictions, fully-automatic systems require to be supervised by a human agent, who corrects the errors. This is a common procedure in the translation industry. This thesis is mainly framed into the machine translation problem, tackled using fully neural systems. Our main objective is to develop more efficient neural machine translation systems, that allow for a more productive usage and deployment of the technology. To this end, we base our contributions on two main cornerstones: how to better use of the system and how to better leverage the data generated along its usage. First, we apply the so-called interactive-predictive framework to neural machine translation. This embeds the human agent and the system into a cooperative correction process, that seeks to reduce the human effort spent for obtaining high-quality translations. We develop different interactive protocols for the neural machine translation technology, namely, a prefix-based and a segment-based protocols. They are implemented by modifying the search space of the model. Moreover, we introduce mechanisms for achieving a fine-grained interaction while maintaining the decoding speed of the system. We carried out a wide experimentation that shows the potential of our contributions. The previous state of the art is overcame by a large margin and the current systems are able to react better to the human interactions. Next, we study how to improve a neural system using the data generated as a byproduct of this correction process. To this end, we rely on two main learning paradigms: online and active learning. Under the first one, the system is updated on the fly, as soon as a sentence is corrected. Hence, the system is continuously learning from the corrections, avoiding previous errors and specializing towards a given user or domain. A large experimentation stressed the adaptive systems under different conditions and domains, demonstrating the capabilities of adaptive systems. Moreover, we also carried out a human evaluation of the system, involving professional users. They were very pleased with the adaptive system, and worked more efficiently using it. The second paradigm, active learning, is devised for the translation of huge amounts of data, that are infeasible to being completely supervised. In this scenario, the system selects samples that are worth to be supervised, and leaves the rest automatically translated. Applying this framework, we obtained reductions of approximately a quarter of the effort required for reaching a desired translation quality. The neural approach also obtained large improvements compared with previous translation technologies. Finally, we address another challenging problem: visual captioning. It consists in generating a description in natural language from a visual object, namely an image or a video. We follow the sequence-to-sequence framework, under a a multimodal perspective. We start by tackling the task of generating captions of videos from a general domain. Next, we move on to a more specific case: describing events from egocentric images, acquired along the day. Since these events are consecutive, we aim to extract inter-eventual relationships, for generating more informed captions. The context-aware model improved the generation quality with respect to a regular one. As final point, we apply the intractive-predictive protocol to these multimodal captioning systems, reducing the effort required for correcting the outputs.
Section 5.4 describes an user evaluation of an adaptive translation system. This was done in collaboration with Miguel Domingo and the company Pangeanic, with funding from the Spanish Center for Technological and Industrial Development (Centro para el Desarrollo Tecnológico Industrial). [...] Most of Chapter 6 is the result of a collaboration with Marc Bolaños, supervised by Prof. Petia Radeva, from Universitat de Barcelona/CVC. This collaboration was supported by the R-MIPRCV network, under grant TIN2014-54728-REDC.
Peris Abril, Á. (2019). Interactivity, Adaptation and Multimodality in Neural Sequence-to-sequence Learning [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/134058
TESIS

This thesis introduces syntax-directed translation based on deep pushdown automata. Necessary theoretical models are introduced in the theoretical part. The most important model, introduced in this thesis, is a deep pushdown transducer. The transducer should be used in syntax analysis, significant part of translation. Practical part consists of an implementation of simple-language interpret based on these models.

Bien qu’ils soient largement utilisés dans le domaine, les mesures vibratoires classiques présentent plusieurs limites. A la base, l’analyse vibratoire ne peut identifier qu’environ 60% des défauts qui peuvent survenir dans les machines. Cependant, les principaux inconvénients des mesures de la vibration sont l’accès difficile au système de transmission afin d’y placer le capteur ainsi que le coût conséquent de la mise en œuvre. Ceci résulte en des problèmes de sensibilité relatifs à la position de l’installation et ceux de difficulté pour distinguer la source de vibration à cause de la diversité des excitations mécaniques qui existent dans l’environnement industriel.Par conséquent, l’analyse des signatures du courant électrique des moteurs s’impose comme une alternative prometteuse à l’analyse vibratoire et a donc fait l’objet d’une attention grandissante au cours des dernières années. En effet, l’analyse des signatures électriques a l’avantage d’être une méthode techniquement accessible, non-intrusive au système et peu coûteuse. Les techniques basées sur le courant et la tension ne requièrent que les mesures électriques du moteur qui sont souvent déjà surveillées pour le contrôle et la protection des machines électriques. Ce processus a été principalement utilisé pour la détection des défauts de moteur tels que la rupture de barres du rotor et les défauts d’excentricité ainsi que les défauts de roulements. En revanche, très peu de recherches concernent la détection des défauts en utilisant l’analyse du courant. En outre, les signaux électriques triphasés sont caractérisés par des représentations géométriques particulières liées à leur forme d’onde qui peuvent servir en tant qu’indicateurs différents offrant des informations supplémentaires. Parmi ces indicateurs géométriques, les transformées de Park et de Concordia modélisent les composantes électriques dans un repère bidimensionnel et toute déviation par rapport à la représentation d’origine indique l’apparition d’un dysfonctionnement. Aussi, les équations différentielles de Frenet-Serret représentent la trajectoire du signal dans un espace euclidien tridimensionnel et indiquent ainsi tout changement dans l’état du système. Bien qu’ils aient été utilisés pour les défauts de roulements, ces indicateurs n’ont pas été appliqués dans la détection des défauts d’engrenages en utilisant l’analyse des signatures des courants électriques. D’où l’idée novatrice de combiner ces indicateurs avec des techniques de traitement de signal, ainsi que des techniques de classification pour le diagnostic des engrenages en utilisant l’analyse des signatures de courant et de tension du moteur électrique.Ainsi, dans ce travail, on propose une nouvelle approche pour le diagnostic des défauts d’engrenages en utilisant l’analyse des courants et des tensions électriques du stator de la machine et ceci en se basant sur un ensemble d’indicateurs géométriques (Transformées de Park et de Concordia ainsi que les propriétés du repère Frenet-Serret). Ces indicateurs font partie d’une bibliothèque de signatures de défauts qui a été construite et qui comprend également les indicateurs classiques utilisés pour un large éventail de défauts. Ainsi, un algorithme combine les acquisitions expérimentales des signaux électriques à des méthodes de traitement de signal avancées (décomposition modale empirique,…). Ensuite, celui-ci sélectionne les indicateurs les plus pertinents au sein de la bibliothèque en se basant sur les algorithmes de sélection de paramètres (sélection séquentielle rétrograde et analyse des composantes principales). Enfin, cette sélection est utilisée pour la classification non-supervisée (K-moyennes) pour la distinction entre l’état sain et l’état défaillant
Although they are widely used, classical vibration measurements have several limitations. Vibration analysis can only identify about 60% of the defects that may occur in mechanical systems. However, the main drawbacks of vibration measurements are the difficult access to the transmission system in order to place the sensor as well as the consequent cost of implementation. This results in sensitivity problems relative to the position of the installation and the difficulty to distinguish the source of vibration because of the diversity of mechanical excitations that exist in the industrial environment.Hence, the Motor Current Signatures Analysis (M.C.S.A.) represents a promising alternative to the vibration analysis and has therefore been the subject of increasing attention in recent years. Indeed, the analysis of electrical signatures has the advantage of being a technically accessible method as well as inexpensive and non-intrusive to the system. Techniques based on currents and voltages only require the motor’s electrical measurements which are often already supervised for the purposes of the control and the protection of the electrical machines. This process was mainly used for the detection of motors faults such as rotor bars breakage and eccentricity faults as well as bearings defects. On the other hand, very little research has been focused on gear faults detection using the current analysis. In addition, three-phase electrical signals are characterized by specific geometric representations related to their waveforms and they can serve as different indicators providing additional information. Among these geometric indicators, the Park and Concordia transforms model the electrical components in a two-dimensional coordinate system and any deviation from the original representation indicates the apparition of a malfunction. Moreover, the differential equations of Frenet-Serret represent the trajectory of the signal in a three-dimensional euclidean space and thus indicate any changes in the state of the system. Although they have been previously used for bearing defects, these indicators have not been applied in the detection of gear defects using the analysis of electrical current signatures. Hence, the innovative idea of combining these indicators with signal processing techniques, as well as classification techniques for gears diagnosis using the three-phase motor’s electrical current signatures analysis is established.Hence, in this work, a new approach is proposed for gear faults diagnosis using the motor currents analysis, based on a set of geometric indicators (Park and Concordia transforms as well as the properties of the Frenet-Serret frame). These indicators are part of a specifically built fault signatures library and which also includes the classical indicators used for a wide range of faults. Thus, a proposed estimation algorithm combines experimental measurements of electrical signals with advanced signal processing methods (Empirical Mode Decomposition, ...). Next, it selects the most relevant indicators within the library based on feature selection algorithms (Sequential Backward Selection and Principal Component Analysis). Finally, this selection is combined with non-supervised classification (K-means) for the distinction between the healthy state and faulty states. It was finally validated with a an additional experimental configuration in different cases with gear faults, bearing faults and combined faults with various load levels

The translation of mRNA into functional proteins is essential for all life. In eukaryotes, aberrant RNAs containing sequence features that stall or severely slow down ribosomes are subject to translation-dependent quality control. Targets include mRNAs encoding a strong secondary structure (No-Go Decay; NGD) or stretches of positively-charged amino acids (Peptide-dependent Translation Arrest/Ribosome Quality Control; PDTA/RQC), mRNAs lacking an in-frame stop codon (Non-Stop Decay; NSD), or defective 18S rRNAs (18S Nonfunctional rRNA Decay; 18S NRD). Previous work from our lab showed that the S. cerevisiae NGD factors DOM34 and HBS1, and PDTA/RQC factor ASC1, all participate in the kinetics of 18S NRD. Upon further investigation of 18S NRD, our research revealed the critical role of ribosomal protein S3 (RPS3), thus adding to the emerging evidence that the ribosome senses its own translational status. While aberrant mRNAs mentioned above can occur endogenously, damaging agents, such as oxidative stress or UV irradiation, can negatively affect the chemical integrity of RNA. Such lesions could lead to translation errors and ribosome stalling. However, current tools to monitor the fate of damaged RNA are quite limited and only provide a low-resolution picture. Therefore, we sought to develop a deep-sequencing method to detect damaged RNA, taking advantage of reverse transcriptase's ability to insert a mutation across a damaged site. Using oxidized RNA as a model damaged RNA, our preliminary data showed increased G>T mutations in oxidized RNA. This method provides the foundation for future work aimed at understanding how cells deal with damaged RNA.

L'apprentissage profond a permis des avancées significatives dans le domaine de la traduction automatique.La traduction automatique neuronale (NMT) s'appuie sur l'entrainement de réseaux de neurones avec un grand nombre de paramètres sur une grand quantité de données parallèles pour apprendre à traduire d'une langue à une autre.Un facteur primordial dans le succès des systèmes NMT est la capacité de concevoir des architectures puissantes et efficaces. Les systèmes de pointe sont des modèles encodeur-décodeurs qui, d'abord, encodent une séquence source sous forme de vecteurs de caractéristiques, puis décodent de façon conditionne la séquence cible.Dans cette thèse, nous remettons en question le paradigme encodeur-décodeur et préconisons de conjointement encoder la source et la cible afin que les deux séquences interagissent à des niveaux d'abstraction croissants. À cette fin, nous introduisons Pervasive Attention, un modèle basé sur des convolutions bidimensionnelles qui encodent conjointement les séquences source et cible avec des interactions qui sont omniprésentes dans le réseau neuronal.Pour améliorer l'efficacité des systèmes NMT, nous étudions la traduction automatique simultanée où la source est lue de manière incrémentielle et le décodeur est alimenté en contextes partiels afin que le modèle puisse alterner entre lecture et écriture. Nous améliorons les agents déterministes qui guident l'alternance lecture / écriture à travers un chemin de décodage rigide et introduisons de nouveaux agents dynamiques pour estimer un chemin de décodage adapté au cas-par-cas.Nous abordons également l'efficacité computationnelle des modèles NMT et affirmons qu'ajouter plus de couches à un réseau de neurones n'est pas requis pour tous les cas.Nous concevons des décodeurs Transformer qui peuvent émettre des prédictions à tout moment dotés de mécanismes d'arrêt adaptatifs pour allouer des ressources en fonction de la complexité de l'instance
In recent years, deep learning has enabled impressive achievements in Machine Translation.Neural Machine Translation (NMT) relies on training deep neural networks with large number of parameters on vast amounts of parallel data to learn how to translate from one language to another.One crucial factor to the success of NMT is the design of new powerful and efficient architectures. State-of-the-art systems are encoder-decoder models that first encode a source sequence into a set of feature vectors and then decode the target sequence conditioning on the source features.In this thesis we question the encoder-decoder paradigm and advocate for an intertwined encoding of the source and target so that the two sequences interact at increasing levels of abstraction. For this purpose, we introduce Pervasive Attention, a model based on two-dimensional convolutions that jointly encode the source and target sequences with interactions that are pervasive throughout the network.To improve the efficiency of NMT systems, we explore online machine translation where the source is read incrementally and the decoder is fed partial contexts so that the model can alternate between reading and writing. We investigate deterministic agents that guide the read/write alternation through a rigid decoding path, and introduce new dynamic agents to estimate a decoding path for each sample.We also address the resource-efficiency of encoder-decoder models and posit that going deeper in a neural network is not required for all instances.We design depth-adaptive Transformer decoders that allow for anytime prediction and sample-adaptive halting mechanisms to favor low cost predictions for low complexity instances and save deeper predictions for complex scenarios

Submitted by SÍLVIO CÉSAR OTERO GARCIA null (silvioce@gmail.com) on 2016-01-29T04:48:49Z No. of bitstreams: 1 Tese.pdf: 4577308 bytes, checksum: 3607cfa97f6ae710923054567d80026c (MD5)
Approved for entry into archive by Sandra Manzano de Almeida (smanzano@marilia.unesp.br) on 2016-01-29T18:35:42Z (GMT) No. of bitstreams: 1 oterogarcia_sc_dr_rcla.pdf: 4577308 bytes, checksum: 3607cfa97f6ae710923054567d80026c (MD5)
Made available in DSpace on 2016-01-29T18:35:42Z (GMT). No. of bitstreams: 1 oterogarcia_sc_dr_rcla.pdf: 4577308 bytes, checksum: 3607cfa97f6ae710923054567d80026c (MD5) Previous issue date: 2015-12-17
Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)
Ce thèse porte sur une traduction et analyse de la thèse de doctorat d’Henri Lebesgue Intégrale, Longueur, Aire. Il s’agit d’une thèse parue en 1902 dans laquelle Lebesgue présente la théorie de la mesure et d’intégration ayant son nom. La traduction, de toute fidélité possible à son original, a pour base la méthodologie de Vinay et Darbelnet (1977). L’analyse se développe à partir du référentiel théorique de l’herméneutique des profondeurs plus spécifiquement, Thompson (2011). Notre intention est de rendre plus accessible une source originale en langue étrangère, le portugais du Brésil dans ce cas, pour contribuer avec les études en Histoire Mathématique mais aussi en avoir comme un outil pédagogique pour l’enseignement des mathématiques.
Neste trabalho traduzimos e analisamos a tese de doutorado de Henri Lebesgue Inté- grale, Longueur, Aire. Publicada em 1902, é nela que Lebesgue apresenta a teoria da medida e integração que levam o seu nome. A tradução, que pretende ser o mais fiel pos- sível ao original, foi feita seguindo a metodologia de Vinay e Darbelnet (1977). A análise foi respaldada pelo referencial teórico da hermenêutica das profundidades proposta por Thompson (2011). O nosso objetivo geral é disponibilizar uma fonte original mais aces- sível para pesquisas em história da matemática bem como facilitar seu uso como recuso pedagógico na educação matemática; trazendo, assim, contribuições para essas áreas do conhecimento.
This thesis deals with the translation and analysis of Henri Lebesgue doctoral thesis, Integrale, Longueur, Aire. Published in 1902, this thesis presents the Lebesgue measure theory and integration named after him. The translation, intended to be as faithful to the original as possible, is based on Vinay and Darbelnet’s (1977) methodology. The analysis is developed from the theoretical reference of the deep hermeneutics, more specifically, Thompson (2011). Our intention is to make available an original source in Portuguese to help with Brazilian studies in History of Mathematics and also ease its use as na educational tool for math teaching.
FAPESP: 2010/18737-1

The design of cultural centre is situated in part Kozia Hôrka( well-known city swimming pool), in its advantage takes natural scenery and calm atmosphere of place. Into action of performance brings a message in form of body of reservoir, function of centre is divided into small parts placed in area Kozia Hôrka. Orientation of objects comes mainly from local natural ispirations. Complex is multifunctional in concept, counts with several sorts of culture - sports events. Whereby the main function of swimming pool is preserved and added for higher comfort of inhabitants. Architecture of objects comes from idea of floating leaf on water level and body of reservoir. Objects stylizely illustrate this idea. The design takes the game of solids of organic and strictly ortogonal shapes. Two mutual opposites, in interaction. Objects smoothy and with respect encroach the environment, which is enough marked by human. Simplicity in used materials ( glass, steel, wood ) give transparency and purity to whole solution.

碩士
國立東華大學
資訊工程學系
106
Language is an indispensable medium for the development of world civilization, but nowadays many minority languages have gradually disappeared and become endangered languages. According to a survey published in 2016 by the Council of Indigenous Peoples of the Republic of China, the lower the average age of Taiwanese indigenous peoples, the lower the vitality of the indigenous languages, which presents a potential crisis of indigenous language loss. In this study, we choose several Taiwanese indigenous ethnic groups with a larger population, the Amis, the Atayal, and the Bunun, building a "Taiwanese Indigenous Languages Translation System." Using the Deep Learning technology, it is able to convert indigenous languages to Mandarin automatically through the Machine Translation after reading the languages. We hope that we can promote the learning of Taiwanese indigenous languages and enhance the inheritance of endangered languages.

abstract: Learning programming involves a variety of complex cognitive activities, from abstract knowledge construction to structural operations, which include program design,modifying, debugging, and documenting tasks. In this work, the objective was to explore and investigate the barriers and obstacles that programming novice learners encountered and how the learners overcome them. Several lab and classroom studies were designed and conducted, the results showed that novice students had different behavior patterns compared to experienced learners, which indicates obstacles encountered. The studies also proved that proper assistance could help novices find helpful materials to read. However, novices still suffered from the lack of background knowledge and the limited cognitive load while learning, which resulted in challenges in understanding programming related materials, especially code examples. Therefore, I further proposed to use the natural language generator (NLG) to generate code explanations for educational purposes. The natural language generator is designed based on Long Short Term Memory (LSTM), a deep-learning translation model. To establish the model, a data set was collected from Amazon Mechanical Turks (AMT) recording explanations from human experts for programming code lines. To evaluate the model, a pilot study was conducted and proved that the readability of the machine generated (MG) explanation was compatible with human explanations, while its accuracy is still not ideal, especially for complicated code lines. Furthermore, a code-example based learning platform was developed to utilize the explanation generating model in programming teaching. To examine the effect of code example explanations on different learners, two lab-class experiments were conducted separately ii in a programming novices’ class and an advanced students’ class. The experiment result indicated that when learning programming concepts, the MG code explanations significantly improved the learning Predictability for novices compared to control group, and the explanations also extended the novices’ learning time by generating more material to read, which potentially lead to a better learning gain. Besides, a completed correlation model was constructed according to the experiment result to illustrate the connections between different factors and the learning effect.
Dissertation/Thesis
Doctoral Dissertation Engineering 2020

碩士
中華大學
資訊工程學系
106
Recently, the deep learning technologies make the conventional vision-based surveillance technologies getting significant improvement in terms of feature discrimination and recognition accuracy, e.g., the vision-based license plate recognition (LPR) technology. However, the conventional LPR systems still face the serious challenges in the outdoor ill-conditioned environments. In this work, we used WebGL to augment the license plate training database required for all-weather adverse environment. In general, the conventional LPR systems consist of the following modules: feature extraction, license plate locating, character segmentation, and character recognition. However, the performances of these module are strongly correlated with some low level image features, e.g., edges, colors, and textures. These low level image features can be influenced significantly by the illumination and view angle variations of license plates such that the recognition accuracy is degraded. Therefore, this project is expected to the following contributions. First, we apply the WebGL to construct the training database of the ill-conditioned outdoor environments. Second, we used the YOLOv2 DNN architecture to develop deep license plate recognition system in the ill-conditioned environments with recognition accuracy 94%.

Ce mémoire s'adresse au problème de traduction de domaine non-supervisée. La traduction non-supervisée cherche à traduire un domaine, le domaine source, à un domaine cible sans supervision. Nous étudions d'abord le problème en utilisant le formalisme du transport optimal. Dans un second temps, nous étudions le problème de transfert de sémantique à haut niveau dans les images en utilisant les avancés en apprentissage de représentations et de transfert d'apprentissages développés dans la communauté d'apprentissage profond. Le premier chapitre est dévoué à couvrir les bases des concepts utilisés dans ce travail. Nous décrivons d'abord l'apprentissage de représentation en incluant la description de réseaux de neurones et de l'apprentissage supervisé et non supervisé. Ensuite, nous introduisons les modèles génératifs et le transport optimal. Nous terminons avec des notions pertinentes sur le transfert d'apprentissages qui seront utiles pour le chapitre 3. Le deuxième chapitre présente \textit{Neural Wasserstein Flow}. Dans ce travail, nous construisons sur la théorie du transport optimal et démontrons que les réseaux de neurones peuvent être utilisés pour apprendre des barycentres de Wasserstein. De plus, nous montrons que les réseaux de neurones peuvent amortir n'importe quel barycentre, permettant d'apprendre une interpolation continue. Nous montrons aussi comment utiliser ces concepts dans le cadre des modèles génératifs. Finalement, nous montrons que notre approche permet d'interpoler des formes et des couleurs. Dans le troisième chapitre, nous nous attaquons au problème de transfert de sémantique haut niveau dans les images. Nous montrons que ceci peut être obtenu simplement avec un GAN conditionné sur la représentation apprise par un réseau de neurone. Nous montrons aussi comment ce processus peut être rendu non-supervisé si la représentation apprise est un regroupement. Finalement, nous montrons que notre approche fonctionne sur la tâche de transfert de MNIST à SVHN. Nous concluons en mettant en relation les deux contributions et proposons des travaux futures dans cette direction.
This thesis is concerned with the problem of unsupervised domain translation. Unsupervised domain translation is the task of transferring one domain, the source domain, to a target domain. We first study this problem using the formalism of optimal transport. Next, we study the problem of high-level semantic image to image translation using advances in representation learning and transfer learning. The first chapter is devoted to reviewing the background concepts used in this work. We first describe representation learning including a description of neural networks and supervised and unsupervised representation learning. We then introduce generative models and optimal transport. We finish with the relevant notions of transfer learning that will be used in chapter 3. The second chapter presents Neural Wasserstein Flow. In this work, we build on the theory of optimal transport and show that deep neural networks can be used to learn a Wasserstein barycenter of distributions. We further show how a neural network can amortize any barycenter yielding a continuous interpolation. We also show how this idea can be used in the generative model framework. Finally, we show results on shape interpolation and colour interpolation. In the third chapter, we tackle the task of high level semantic image to image translation. We show that high level semantic image to image translation can be achieved by simply learning a conditional GAN with the representation learned from a neural network. We further show that we can make this process unsupervised if the representation learning is a clustering. Finally, we show that our approach works on the task of MNIST to SVHN.

Dans ce mémoire, nous examinons certaines propriétés des représentations distribuées de mots et nous proposons une technique pour élargir le vocabulaire des systèmes de traduction automatique neurale. En premier lieu, nous considérons un problème de résolution d'analogies bien connu et examinons l'effet de poids adaptés à la position, le choix de la fonction de combinaison et l'impact de l'apprentissage supervisé. Nous enchaînons en montrant que des représentations distribuées simples basées sur la traduction peuvent atteindre ou dépasser l'état de l'art sur le test de détection de synonymes TOEFL et sur le récent étalon-or SimLex-999. Finalament, motivé par d'impressionnants résultats obtenus avec des représentations distribuées issues de systèmes de traduction neurale à petit vocabulaire (30 000 mots), nous présentons une approche compatible à l'utilisation de cartes graphiques pour augmenter la taille du vocabulaire par plus d'un ordre de magnitude. Bien qu'originalement développée seulement pour obtenir les représentations distribuées, nous montrons que cette technique fonctionne plutôt bien sur des tâches de traduction, en particulier de l'anglais vers le français (WMT'14).
In this thesis, we examine some properties of word embeddings and propose a technique to handle large vocabularies in neural machine translation. We first look at a well-known analogy task and examine the effect of position-dependent weights, the choice of combination function and the impact of supervised learning. We then show that simple embeddings learnt with translational contexts can match or surpass the state of the art on the TOEFL synonym detection task and on the recently introduced SimLex-999 word similarity gold standard. Finally, motivated by impressive results obtained by small-vocabulary (30,000 words) neural machine translation embeddings on some word similarity tasks, we present a GPU-friendly approach to increase the vocabulary size by more than an order of magnitude. Despite originally being developed for obtaining the embeddings only, we show that this technique actually works quite well on actual translation tasks, especially for English to French (WMT'14).

Title: Indonesian-English Neural Machine Translation Author: Meisyarah Dwiastuti Department: Institute of Formal and Applied Linguistics Supervisor: Mgr. Martin Popel, Ph.D., Institute of Formal and Applied Linguis- tics Abstract: In this thesis, we conduct a study on neural machine translation (NMT) for an under-studied language, Indonesian, specifically for English-Indonesian (EN-ID) and Indonesian-English (ID-EN) in a low-resource domain, TED talks. Our goal is to implement domain adaptation methods to improve the low-resource EN-ID and ID-EN NMT systems. First, we implement model fine-tuning method for EN-ID and ID-EN NMT systems by leveraging a large parallel corpus contain- ing movie subtitles. Our analysis shows the benefit of this method for the improve- ment of both systems. Second, we improve our ID-EN NMT system by leveraging English monolingual corpora through back-translation. Our back-translation ex- periments focus on how to incorporate the back-translated monolingual corpora to the training set, in which we investigate various existing training regimes and introduce a novel 4-way-concat training regime. We also analyze the effect of fine- tuning our back-translation models with different scenarios. Experimental results show that our method of implementing back-translation followed by model...

Machine Translation Using Syntactic Analysis Martin Popel This thesis describes our improvement of machine translation (MT), with a special focus on the English-Czech language pair, but using techniques ap- plicable also to other languages. First, we present multiple improvements of the deep-syntactic system TectoMT. For instance, we implemented a novel context-sensitive translation model, comparing several machine learning ap- proaches. We also adapted TectoMT to other domains and languages. Sec- ond, we present Transformer - a state-of-the-art end-to-end neural MT sys- tem. We analyzed in detail the effect of several training hyper-parameters. With our optimized training, the system outperformed the best result on the WMT2017 test set by +1.0 BLEU. We further extended this system by uti- lization of monolingual training data and by a new type of backtranslation (+2.8 BLEU compared to the baseline system). In addition, we leveraged domain adaptation and the effect of "translationese" (i.e which language in parallel data is the original and which is the translation) to optimize MT systems for original-language and translated-language data (gaining further +0.2 BLEU). Our improved neural MT system significantly (p¡0.05) out- performed all other systems in English-Czech and Czech-English WMT2018 shared tasks,...

La mémoire de traduction (MT) joue un rôle décisif lors de la traduction et constitue une base de données idéale pour la plupart des professionnels de la langue. Cependant, une MT est très sujète au bruit et, en outre, il n’y a pas de source spécifique. Des efforts importants ont été déployés pour nettoyer des MT, en particulier pour former un meilleur système de traduction automatique. Dans cette thèse, nous essayons également de nettoyer la MT mais avec un objectif plus large : maintenir sa qualité globale et la rendre suffisament robuste pour un usage interne dans les institutions. Nous proposons un processus en deux étapes : d’abord nettoyer une MT institutionnelle (presque propre), c’est-à-dire éliminer le bruit, puis détecter les textes traduits à partir de systèmes neuronaux de traduction. Pour la tâche d’élimination du bruit, nous proposons une architecture impliquant cinq approches basées sur l’heuristique, l’ingénierie fonctionnelle et l’apprentissage profond. Nous évaluons cette tâche à la fois par annotation manuelle et traduction automatique (TA). Nous signalons un gain notable de +1,08 score BLEU par rapport à un système de nettoyage état de l’art. Nous proposons également un outil Web qui annote automatiquement les traductions incorrectes, y compris mal alignées, pour les institutions afin de maintenir une MT sans erreur. Les modèles neuronaux profonds ont considérablement amélioré les systèmes MT, et ces systèmes traduisent une immense quantité de texte chaque jour. Le matériel traduit par de tels systèmes finissent par peuplet les MT, et le stockage de ces unités de traduction dans TM n’est pas idéal. Nous proposons un module de détection sous deux conditions: une tâche bilingue et une monolingue (pour ce dernier cas, le classificateur ne regarde que la traduction, pas la phrase originale). Nous rapportons une précision moyenne d’environ 85 % en domaine et 75 % hors domaine dans le cas bilingue et 81 % en domaine et 63 % hors domaine pour le cas monolingue en utilisant des classificateurs d’apprentissage profond.
Translation Memory (TM) plays a decisive role during translation and is the go-to database for most language professionals. However, they are highly prone to noise, and additionally, there is no one specific source. There have been many significant efforts in cleaning the TM, especially for training a better Machine Translation system. In this thesis, we also try to clean the TM but with a broader goal of maintaining its overall quality and making it robust for internal use in institutions. We propose a two-step process, first clean an almost clean TM, i.e. noise removal and then detect texts translated from neural machine translation systems. For the noise removal task, we propose an architecture involving five approaches based on heuristics, feature engineering, and deep-learning and evaluate this task by both manual annotation and Machine Translation (MT). We report a notable gain of +1.08 BLEU score over a state-of-the-art, off-the-shelf TM cleaning system. We also propose a web-based tool “OSTI: An Open-Source Translation-memory Instrument” that automatically annotates the incorrect translations (including misaligned) for the institutions to maintain an error-free TM. Deep neural models tremendously improved MT systems, and these systems are translating an immense amount of text every day. The automatically translated text finds a way to TM, and storing these translation units in TM is not ideal. We propose a detection module under two settings: a monolingual task, in which the classifier only looks at the translation; and a bilingual task, in which the source text is also taken into consideration. We report a mean accuracy of around 85% in-domain and 75% out-of-domain for bilingual and 81% in-domain and 63% out-of-domain from monolingual tasks using deep-learning classifiers.

Multimodality in Machine Translation Jindřich Libovický Traditionally, most natural language processing tasks are solved within the lan- guage, relying on distributional properties of words. Representation learning abilities of deep learning recently allowed using additional information source by grounding the representations in the visual modality. One of the tasks that attempt to exploit the visual information is multimodal machine translation: translation of image captions when having access to the original image. The thesis summarizes joint processing of language and real-world images using deep learning. It gives an overview of the state of the art in multimodal machine translation and describes our original contribution to solving this task. We introduce methods of combining multiple inputs of possibly different modalities in recurrent and self-attentive sequence-to-sequence models and show results on multimodal machine translation and other tasks related to machine translation. Finally, we analyze how the multimodality influences the semantic properties of the sentence representation learned by the networks and how that relates to translation quality.