Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Since its introduction in 2014 [Goodfellow et al., 2014], the architecture of Generative Adversarial Networks (GANs) have experienced various evolutions to reach its current state where it is capable to recreate realistic images of any given context. Those improvements, both in terms of complexity and stability, enabled successful applications of GANs frameworks in the field of computer vision and transfer learning. On the other hand, GANs lack of successful applications within the field of Natural Language Processing (NLP) where models based on Transformers architecture, such as Bidirectional Encoder Representations from Transformers (BERT) and Generative Pre-Training (GPT), remain the current state-of-the-art for various NLP tasks.

Given this current situation, this thesis investigates why GANs remain underused for NLP tasks. As such, we explore some researchers’ proposals within the area of Dialog Systems by using data from the Daily Dialog dataset, a human-written and multi-turn dialog set reflecting daily human communication.

Moreover, we investigate the influence of an embedding layer of the proposed GAN models. In order to do so first, we test pre-trained “word-level” embeddings, such as Stanford's Glove and Spacy embeddings. 

Second, we train the model by using our own word embeddings coming from the Daily Dialog dataset. The Word2Vec algorithm is used in this case. Third, we explore the idea of using BERT as a contextualized word embeddings. From these experiments it was observed that the use of pre-trained embeddings, not only accelerates the convergence during the training but also, improves the quality of the produced samples by the model, to some extents avoiding an early arrival of mode collapse.

In conclusion, despite their limited success in the NLP area, GAN-trained models offer an interesting approach during the training phase, as the generator G is able to produce different but potentially correct response samples and is not penalized by not producing the most likely single correct sequence of words. This actually follows an important characteristic of the human learning process. Overall, this thesis successfully explores propositions made to tackle drawbacks of the GAN architecture within the NLP area and opens doors for critical progresses in the area.

machine learning --- gan --- auto-encoders --- daily dialog --- deep learning --- Ingénierie, informatique & technologie > Sciences informatiques

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

This book includes 13 papers published in Special Issue ("Visual and Camera Sensors") of the journal Sensors. The goal of this Special Issue was to invite high-quality, state-of-the-art research papers dealing with challenging issues in visual and camera sensors.

self-assembly device --- 3D point clouds --- accuracy analysis --- VSLAM-photogrammetric algorithm --- portable mobile mapping system --- low-cost device --- BIM --- camera calibration --- DLT --- PnP --- weighted DLT --- uncertainty --- covariance --- robustness --- visual-inertial --- semi-direct SLAM --- multi-sensor fusion --- side-rear-view monitoring system --- automatic online calibration --- Hough-space --- unmanned aerial vehicle --- autonomous landing --- deep-learning-based motion deblurring and marker detection --- network slimming --- pruning model --- convolutional neural network --- convolutional filter --- classification --- multimodal human recognition --- blur image restoration --- DeblurGAN --- CNN --- facial expression recognition system --- computer vision --- multi-scale featured local binary pattern --- unsharp masking --- machine learning --- lens distortion --- DoF-dependent --- distortion partition --- vision measurement --- pathological site classification --- in vivo endoscopy --- computer-aided diagnosis --- artificial intelligence --- ensemble learning --- convolutional auto-encoders --- local image patch --- point pair feature --- plank recognition --- robotic grasping --- flying object detection --- drone --- image processing --- camera networks --- open-pit mine slope monitoring --- optimum deployment --- close range photogrammetry --- three-dimensional reconstruction --- OCD4M

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

This book is a collection of published articles from the Sensors Special Issue on "Convergence of Intelligent Data Acquisition and Advanced Computing Systems". It includes extended versions of the conference contributions from the 10th IEEE International Conference on Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications (IDAACS’2019), Metz, France, as well as external contributions.

Technology: general issues --- Energy industries & utilities --- automotive --- current --- electric power train --- electric vehicle --- embedded systems --- delay --- detection --- distributed systems --- measurements --- power train --- sensor --- signals --- time delay estimation --- unmanned aerial vehicles --- wireless sensor networks --- intelligent data processing --- trajectory planning --- relevant data extraction --- data consensus --- Internet of Things --- precision agriculture --- system identification --- smart building --- artificial neural network --- energy efficiency --- black box modeling --- educational robotics --- data acquisition --- sensors --- ROS --- STEM --- CNN (Convolutional neural networks) --- deep learning --- pavement defects --- residual connection --- attention gate --- atrous spatial pyramid pooling --- intelligent charging --- demand response --- linear programming --- optimization --- smart parking --- smart grid --- ODE Solver --- OpenCL --- Parareal --- parallel/multi-core computing --- sensing systems --- heterogenous embedded systems --- deep sparse auto-encoders --- medical diagnosis --- linear model --- data classification --- PSO algorithm --- safety-related system --- component --- FPGA-designing --- logical and power-oriented checkability --- hidden faults --- clock signal --- consumed and dissipated power --- temperature and current consumption sensors --- automotive --- current --- electric power train --- electric vehicle --- embedded systems --- delay --- detection --- distributed systems --- measurements --- power train --- sensor --- signals --- time delay estimation --- unmanned aerial vehicles --- wireless sensor networks --- intelligent data processing --- trajectory planning --- relevant data extraction --- data consensus --- Internet of Things --- precision agriculture --- system identification --- smart building --- artificial neural network --- energy efficiency --- black box modeling --- educational robotics --- data acquisition --- sensors --- ROS --- STEM --- CNN (Convolutional neural networks) --- deep learning --- pavement defects --- residual connection --- attention gate --- atrous spatial pyramid pooling --- intelligent charging --- demand response --- linear programming --- optimization --- smart parking --- smart grid --- ODE Solver --- OpenCL --- Parareal --- parallel/multi-core computing --- sensing systems --- heterogenous embedded systems --- deep sparse auto-encoders --- medical diagnosis --- linear model --- data classification --- PSO algorithm --- safety-related system --- component --- FPGA-designing --- logical and power-oriented checkability --- hidden faults --- clock signal --- consumed and dissipated power --- temperature and current consumption sensors

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

In many machine learning applications, available datasets are sometimes incomplete, noisy or affected by artifacts. In supervised scenarios, it could happen that label information has low quality, which might include unbalanced training sets, noisy labels and other problems. Moreover, in practice, it is very common that available data samples are not enough to derive useful supervised or unsupervised classifiers. All these issues are commonly referred to as the low-quality data problem. This book collects novel contributions on machine learning methods for low-quality datasets, to contribute to the dissemination of new ideas to solve this challenging problem, and to provide clear examples of application in real scenarios.

Information technology industries --- open contours --- similarly shaped fish species --- Discrete Cosine Transform (DCT) --- Discrete Fourier Transform (DFT) --- Extreme Learning Machines (ELM) --- feature engineering --- small data-sets --- optimization --- machine learning --- preprocessing --- image generation --- weighted interpolation map --- binarization --- single sample per person --- root canal measurement --- multifrequency impedance --- data augmentation --- neural network --- functional magnetic resonance imaging --- independent component analysis --- deep learning --- recurrent neural network --- functional connectivity --- episodic memory --- small sample learning --- feature selection --- noise elimination --- space consistency --- label correlations --- empirical mode decomposition --- sparse representations --- tensor decomposition --- tensor completion --- machine translation --- pairwise evaluation --- educational data --- small datasets --- noisy datasets --- smart building --- Internet of Things (IoT) --- Markov Chain Monte Carlo (MCMC) --- ontology --- graph model --- Artificial Neural Network --- Discriminant Analysis --- dengue --- feature extraction --- sound event detection --- non-negative matrix factorization --- ultrasound images --- shadow detection --- shadow estimation --- auto-encoders --- semi-supervised learning --- prediction --- feature importance --- feature elimination --- hierarchical clustering --- Parkinson’s disease --- few-shot learning --- permutation-variable importance --- topological data analysis --- persistent entropy --- support-vector machine --- data science --- intelligent decision support --- social vulnerability --- gender-gap --- digital-gap --- COVID19 --- policy-making support --- artificial intelligence --- imperfect dataset --- open contours --- similarly shaped fish species --- Discrete Cosine Transform (DCT) --- Discrete Fourier Transform (DFT) --- Extreme Learning Machines (ELM) --- feature engineering --- small data-sets --- optimization --- machine learning --- preprocessing --- image generation --- weighted interpolation map --- binarization --- single sample per person --- root canal measurement --- multifrequency impedance --- data augmentation --- neural network --- functional magnetic resonance imaging --- independent component analysis --- deep learning --- recurrent neural network --- functional connectivity --- episodic memory --- small sample learning --- feature selection --- noise elimination --- space consistency --- label correlations --- empirical mode decomposition --- sparse representations --- tensor decomposition --- tensor completion --- machine translation --- pairwise evaluation --- educational data --- small datasets --- noisy datasets --- smart building --- Internet of Things (IoT) --- Markov Chain Monte Carlo (MCMC) --- ontology --- graph model --- Artificial Neural Network --- Discriminant Analysis --- dengue --- feature extraction --- sound event detection --- non-negative matrix factorization --- ultrasound images --- shadow detection --- shadow estimation --- auto-encoders --- semi-supervised learning --- prediction --- feature importance --- feature elimination --- hierarchical clustering --- Parkinson’s disease --- few-shot learning --- permutation-variable importance --- topological data analysis --- persistent entropy --- support-vector machine --- data science --- intelligent decision support --- social vulnerability --- gender-gap --- digital-gap --- COVID19 --- policy-making support --- artificial intelligence --- imperfect dataset

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

In many machine learning applications, available datasets are sometimes incomplete, noisy or affected by artifacts. In supervised scenarios, it could happen that label information has low quality, which might include unbalanced training sets, noisy labels and other problems. Moreover, in practice, it is very common that available data samples are not enough to derive useful supervised or unsupervised classifiers. All these issues are commonly referred to as the low-quality data problem. This book collects novel contributions on machine learning methods for low-quality datasets, to contribute to the dissemination of new ideas to solve this challenging problem, and to provide clear examples of application in real scenarios.

open contours --- similarly shaped fish species --- Discrete Cosine Transform (DCT) --- Discrete Fourier Transform (DFT) --- Extreme Learning Machines (ELM) --- feature engineering --- small data-sets --- optimization --- machine learning --- preprocessing --- image generation --- weighted interpolation map --- binarization --- single sample per person --- root canal measurement --- multifrequency impedance --- data augmentation --- neural network --- functional magnetic resonance imaging --- independent component analysis --- deep learning --- recurrent neural network --- functional connectivity --- episodic memory --- small sample learning --- feature selection --- noise elimination --- space consistency --- label correlations --- empirical mode decomposition --- sparse representations --- tensor decomposition --- tensor completion --- machine translation --- pairwise evaluation --- educational data --- small datasets --- noisy datasets --- smart building --- Internet of Things (IoT) --- Markov Chain Monte Carlo (MCMC) --- ontology --- graph model --- Artificial Neural Network --- Discriminant Analysis --- dengue --- feature extraction --- sound event detection --- non-negative matrix factorization --- ultrasound images --- shadow detection --- shadow estimation --- auto-encoders --- semi-supervised learning --- prediction --- feature importance --- feature elimination --- hierarchical clustering --- Parkinson’s disease --- few-shot learning --- permutation-variable importance --- topological data analysis --- persistent entropy --- support-vector machine --- data science --- intelligent decision support --- social vulnerability --- gender-gap --- digital-gap --- COVID19 --- policy-making support --- artificial intelligence --- imperfect dataset