This paper proposes an utterance veri cation system for hidden Markov model (HMM) based automatic speech recognition systems. A veri cation objective function, based on a multi-layer-perceptron (MLP), is adopted which combines con dence measures from both the recognition and veri cation models. Discriminative minimum veri cation error training is applied for optimizing the parameters of the MLP...

In this paper, we propose a novel method for combining deep learning and classical feature based models using a Multi-Layer Perceptron (MLP) network for financial sentiment analysis. We develop various deep learning models based on Convolutional Neural Network (CNN), Long Short Term Memory (LSTM) and Gated Recurrent Unit (GRU). These are trained on top of pre-trained, autoencoder-based, financi...

The time series of wind power is influenced by many external factors, showing strong volatility and randomness. Aiming at the problem low prediction accuracy series, this paper proposes a short-term framework based on two-layer decomposition combination ensemble model deep network, which composed complete empirical mode (CEEMD), sample entropy (SE), stacking ensemble, linear regression (LR), va...

Using richly parameterised models for small datasets can be justified from a theoretical point of view according to some results due to Bartlett [1] which show that the generalization performance of a multi layer perceptron (MLP) depends more on the L1 norm ‖c‖1 of the weights between the hidden and the output layer rather than on the number of parameters in the model. In this paper we investig...

Phrase-based statistical machine translation (PBSMT) decoders translate source sentences one phrase at a time using strong independence assumptions over the source phrases. Translation table scores are typically independent of context, language model scores depend on a few words surrounding the target phrase and distortion models do not influence directly the choice of target phrases. In this w...

This paper proposes the application to the liver fibrosis stadialization of a novel training technique of feed-forward neural networks based on the Bayesian paradigm. Using the Pearson’s r correlation coefficient instead of the standard backpropagation algorithm to update the synaptic weights of a multi-layer perceptron, the proposed model is compared with traditional machine learning algorithm...

-------------------------------------------------------------------ABSTRACT---------------------------------------------------------------Prediction of rainfall for a region is of utmost importance for planning, design and management of irrigation and drainage systems. This can be achieved by different approaches such as deterministic, conceptual, stochastic and Artificial Neural Network (ANN)....