From Word Embeddings to Large Vocabulary Neural Machine Translation

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).

Neural changes associated with semantic processing in healthy aging despite intact behavioral performance

Semantic memory recruits an extensive neural network including the left inferior prefrontal cortex (IPC) and the left temporoparietal region, which are involved in semantic control processes, as well as the anterior temporal lobe region (ATL) which is considered to be involved in processing semantic information at a central level. However, little is known about the underlying neuronal integrity of the semantic network in normal aging. Young and older healthy adults carried out a semantic judgment task while their cortical activity was recorded using magnetoencephalography (MEG). Despite equivalent behavioral performance, young adults activated the left IPC to a greater extent than older adults, while the latter group recruited the temporoparietal region bilaterally and the left ATL to a greater extent than younger adults. Results indicate that significant neuronal changes occur in normal aging, mainly in regions underlying semantic control processes, despite an apparent stability in performance at the behavioral level.

Metabolomics analysis in rats with thiamine deficiency identifies key metabolites in vulnerable brain regions and suggests neural stem progenitor cells play a role in ameliorating metabolic dysfunction

La documentation scientifique fait état de la présence, chez l’adulte, de cellules souches et progénitrices neurales (CSPN) endogènes dans les zones sous-ventriculaire et sous-granulaire du cerveau ainsi que dans le gyrus denté de l’hippocampe. De plus, un postulat selon lequel il serait également possible de retrouver ce type de cellules dans la moelle épinière et le néocortex des mammifères adultes a été énoncé. L’encéphalopathie de Wernicke, un trouble neurologique grave toutefois réversible qui entraîne un dysfonctionnement, voire une défaillance du cerveau, est causée principalement par une carence importante en thiamine (CT). Des observations récentes laissent envisager que les facteurs en cause dans la prolifération et la différenciation des CSPN pourraient également jouer un rôle important lors d’un épisode de CT. L’hypothèse, selon laquelle l’identification de nouveaux métabolites entrant dans le mécanisme ou la séquence de réactions se soldant en une CT pourraient en faciliter la compréhension, a été émise au moyen d'une démarche en cours permettant d’établir le profil des modifications métaboliques qui surviennent en de telles situations. Cette approche a été utilisée pour constater les changements métaboliques survenus au niveau du foyer cérébral dans un modèle de rats déficients en thiamine (rats DT), particulièrement au niveau du thalamus et du colliculus inférieur (CI). La greffe de CSPN a quant à elle été envisagée afin d’apporter de nouvelles informations sur la participation des CSPN lors d’un épisode de CT et de déterminer les bénéfices thérapeutiques potentiels offerts par cette intervention. Les sujets de l’étude étaient répartis en quatre groupes expérimentaux : un premier groupe constitué de rats dont la CT était induite par la pyrithiamine (rats DTiP), un deuxième groupe constitué de rats-contrôles nourris ensemble (« pair-fed control rats » ou rats PFC) ainsi que deux groupes de rats ayant subi une greffe de CSPN, soit un groupe de rats DTiP greffés et un dernier groupe constitué de rats-contrôles (rats PFC) greffés. Les échantillons de foyers cérébraux (thalamus et CI) des quatre groupes de rats ont été prélevés et soumis à des analyses métabolomiques non ciblées ainsi qu’à une analyse visuelle par microscopie à balayage électronique (SEM). Une variété de métabolites-clés a été observée chez les groupes de rats déficients en thiamine (rats DTiP) en plus de plusieurs métabolites dont la documentation ne faisait pas mention. On a notamment constaté la présence d’acides biliaires, d’acide cynurénique et d’acide 1,9— diméthylurique dans le thalamus, alors que la présence de taurine et de carnosine a été observée dans le colliculus inférieur. L’étude a de plus démontré une possible implication des CSPN endogènes dans les foyers cérébraux du thalamus et du colliculus inférieur en identifiant les métabolites-clés ciblant les CSPN. Enfin, les analyses par SEM ont montré une amélioration notable des tissus à la suite de la greffe de CSPN. Ces constatations suggèrent que l’utilisation de CSPN pourrait s’avérer une avenue thérapeutique intéressante pour soulager la dégénérescence symptomatique liée à une grave carence en thiamine chez l’humain.

Tunnel-Based Artificial Neural Network Technique to Calculate the Resonant Frequency of a Thick-Substrate Microstrip Antenna

The mathematical formulation of empirically developed formulas Jirr the calculation of the resonant frequency of a thick-substrate (h s 0.08151 A,,) microstrip antenna has been analyzed. With the use qt' tunnel-based artificial neural networks (ANNs), the resonant frequency of antennas with h satisfying the thick-substrate condition are calculated and compared with the existing experimental results and also with the simulation results obtained with the use of an IE3D software package. The artificial neural network results are in very good agreement with the experimental results

Comparison Study of Pattern Synthesis Techniques Using Neural Networks

In this paper, a comparison study among three neuralnetwork algorithms for the synthesis of array patterns is presented. The neural networks are used to estimate the array elements' excitations for an arbitrary pattern. The architecture of the neural networks is discussed and simulation results are presented. Two new neural networks, based on radial basis functions (RBFs) and wavelet neural networks (WNNs), are introduced. The proposed networks offer a more efficient synthesis procedure, as compared to other available techniques

AN ARTIFICIAL NEURAL NETWORK FOR SONAR TARGET DETECTION

Neural Network has emerged as the topic of the day. The spectrum of its application is as wide as from ECG noise filtering to seismic data analysis and from elementary particle detection to electronic music composition. The focal point of the proposed work is an application of a massively parallel connectionist model network for detection of a sonar target. This task is segmented into: (i) generation of training patterns from sea noise that contains radiated noise of a target, for teaching the network;(ii) selection of suitable network topology and learning algorithm and (iii) training of the network and its subsequent testing where the network detects, in unknown patterns applied to it, the presence of the features it has already learned in. A three-layer perceptron using backpropagation learning is initially subjected to a recursive training with example patterns (derived from sea ambient noise with and without the radiated noise of a target). On every presentation, the error in the output of the network is propagated back and the weights and the bias associated with each neuron in the network are modified in proportion to this error measure. During this iterative process, the network converges and extracts the target features which get encoded into its generalized weights and biases.In every unknown pattern that the converged network subsequently confronts with, it searches for the features already learned and outputs an indication for their presence or absence. This capability for target detection is exhibited by the response of the network to various test patterns presented to it.Three network topologies are tried with two variants of backpropagation learning and a grading of the performance of each combination is subsequently made.

Studies on the preparation of Supported nickel catalysts using bis(ethylenediamine)nickel(II) complexes as precursors

Dept.of Applied Chemistry,Cochin University of Science and Technology

Neural Network based studies on Spectroscopic Analysis and Image Processing

International School of Photonics, Cochin University of Science and Technology

Identification of spectral lines of elements using artificial neural networks

Artificial neural networks (ANNs) are relatively new computational tools that have found extensive utilization in solving many complex real-world problems. This paper describes how an ANN can be used to identify the spectral lines of elements. The spectral lines of Cadmium (Cd), Calcium (Ca), Iron (Fe), Lithium (Li), Mercury (Hg), Potassium (K) and Strontium (Sr) in the visible range are chosen for the investigation. One of the unique features of this technique is that it uses the whole spectrum in the visible range instead of individual spectral lines. The spectrum of a sample taken with a spectrometer contains both original peaks and spurious peaks. It is a tedious task to identify these peaks to determine the elements present in the sample. ANNs capability of retrieving original data from noisy spectrum is also explored in this paper. The importance of the need of sufficient data for training ANNs to get accurate results is also emphasized. Two networks are examined: one trained in all spectral lines and other with the persistent lines only. The network trained in all spectral lines is found to be superior in analyzing the spectrum even in a noisy environment.

Invariant Characterization of Neural Systems

It is shown that the invariant integral, viz., the Kolmogorov second entropy, is eminently suited to characterize EEG quantitatively. The estimation obtained for a "clinically normal" brain is compared with a previous result obtained from the EEG of a person under epileptic seizure.

Document Image Segmentation and Compression using Artificial Neural Networks and Evolutionary Methods

The work is intended to study the following important aspects of document image processing and develop new methods. (1) Segmentation ofdocument images using adaptive interval valued neuro-fuzzy method. (2) Improving the segmentation procedure using Simulated Annealing technique. (3) Development of optimized compression algorithms using Genetic Algorithm and parallel Genetic Algorithm (4) Feature extraction of document images (5) Development of IV fuzzy rules. This work also helps for feature extraction and foreground and background identification. The proposed work incorporates Evolutionary and hybrid methods for segmentation and compression of document images. A study of different neural networks used in image processing, the study of developments in the area of fuzzy logic etc is carried out in this work

MicroRNA–mRNA interaction network using TSK-type recurrent neural fuzzy network

MicroRNAs are short non-coding RNAs that can regulate gene expression during various crucial cell processes such as differentiation, proliferation and apoptosis. Changes in expression profiles of miRNA play an important role in the development of many cancers, including CRC. Therefore, the identification of cancer related miRNAs and their target genes are important for cancer biology research. In this paper, we applied TSK-type recurrent neural fuzzy network (TRNFN) to infer miRNA–mRNA association network from paired miRNA, mRNA expression profiles of CRC patients. We demonstrated that the method we proposed achieved good performance in recovering known experimentally verified miRNA–mRNA associations. Moreover, our approach proved successful in identifying 17 validated cancer miRNAs which are directly involved in the CRC related pathways. Targeting such miRNAs may help not only to prevent the recurrence of disease but also to control the growth of advanced metastatic tumors. Our regulatory modules provide valuable insights into the pathogenesis of cancer

Face Recognition using Probabilistic Neural Networks

In this paper we address the problem of face detection and recognition of grey scale frontal view images. We propose a face recognition system based on probabilistic neural networks (PNN) architecture. The system is implemented using voronoi/ delaunay tessellations and template matching. Images are segmented successfully into homogeneous regions by virtue of voronoi diagram properties. Face verification is achieved using matching scores computed by correlating edge gradients of reference images. The advantage of classification using PNN models is its short training time. The correlation based template matching guarantees good classification results

Face Recognition using Probabilistic Neural Networks

n this paper we address the problem of face detection and recognition of grey scale frontal view images. We propose a face recognition system based on probabilistic neural networks (PNN) architecture. The system is implemented using voronoi/ delaunay tessellations and template matching. Images are segmented successfully into homogeneous regions by virtue of voronoi diagram properties. Face verification is achieved using matching scores computed by correlating edge gradients of reference images. The advantage of classification using PNN models is its short training time. The correlation based template matching guarantees good classification results.

Using Neural Network Classifier Support Vector Machine Regression for the prediction of Melting Point of Drug – like compounds

In our study we use a kernel based classification technique, Support Vector Machine Regression for predicting the Melting Point of Drug – like compounds in terms of Topological Descriptors, Topological Charge Indices, Connectivity Indices and 2D Auto Correlations. The Machine Learning model was designed, trained and tested using a dataset of 100 compounds and it was found that an SVMReg model with RBF Kernel could predict the Melting Point with a mean absolute error 15.5854 and Root Mean Squared Error 19.7576