The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this study we introduce a new approach to train a fully recurrent artificial neural network by solving a constraint satisfaction problem using the quotient gradient method. The quotient gradient method is a trajectory based methodology for global optimization that does not suffer from the problem of local minima encountered in Newton based methods. Simulation results show that the network trained...
Carefully injected noise can speed the convergence and accuracy of video classification with recurrent backpropagation (RBP). This noise-boost uses the recent results that backpropagation is a special case of the generalized expectation maximization (EM) algorithm and that careful noise injection can always speed the average convergence of the EM algorithm to a local maximum of the log-likelihood...
The paper proposes a new method for improving the performance of Recurrent Neural Networks. The proposed method uses two parallel recurrent layers which execute independent of each other. The final output of recurrent layer at any time step is computed as the mean of the modulus of the output of these two layers. The proposed method attempts to overcome the limitations of the existing Recurrent Neural...
In this work we propose a new framework for combined feature extraction and classification. The base idea stems from the sparse representation based classification; where in the training samples from each class are assumed to form a basis for representing the same. Later studies learned a basis for each class using dictionary learning; these were shallow techniques where only one level of dictionary...
The method presented extends a given regression neural network to make its performance improve. The modification affects the learning procedure only, hence the extension may be easily omitted during evaluation without any change in prediction. It means that the modified model may be evaluated as quickly as the original one but tends to perform better. This improvement is possible because the modification...
Word embeddings are a low-dimensional vector representation of words that incorporates context. TWo popular methods are word2vec and global vectors (GloVe). Word2vec is a single-hidden layer feedforward neural network (SLFN) that has an auto-encoder influence for computing a word context matrix using backpropagation for training. GloVe computes the word context matrix first then performs matrix factorization...
Next to stroke, the epilepsy is one of the most serious neurological disorders. Due to the hyperactive firing of neurons on a cellular level, epilepsy is caused. The activities of the cortical regions are recorded with the help of Electroencephalogram (EEG) which helps in the diagnosis of epilepsy. The normal patterns of the activities of neurons becomes severely disturbed in the case of epilepsy,...
The purpose of this research was to optimize the backpropagation algorithm process by adding the Nguyen-Widrow method in input layer of feed-forward process and adapting the learning rate parameter in backward process in the backpropagation. In the preprocessing usually the data have not been normalized so the significant to the target output need to be reduce in the input layer process [1]. By embedded...
Training deep recurrent neural network (RNN) architectures is complicated due to the increased network complexity. This disrupts the learning of higher order abstracts using deep RNN. In case of feed-forward networks training deep structures is simple and faster while learning long-term temporal information is not possible. In this paper we propose a residual memory neural network (RMN) architecture...
This paper presents the ideal approach to how to minimize the time taken by reinforcement learning to train the model. Similar to Computer vision the progress in reinforcement learning is not influenced by new ideas but mostly by the computation, large data, infrastructure and efficiency of algorithm. These 4 things only influenced the reinforcement learning RL model. How much time it will take to...
We present Deep Sparse-coded Network (DSN), a deep architecture based on multilayer sparse coding. It has been considered difficult to learn a useful feature hierarchy by stacking sparse coding layers in a straightforward manner. The primary reason is the modeling assumption for sparse coding that takes in a dense input and yields a sparse output vector. Applying a sparse coding layer on the output...
Recently deep learning methods have been applied to image super-resolution (SR). Typically, these approaches involve training a single convolutional neural network that is trained to perform resolution enhancement. We propose a new low-complexity but effective algorithm called Superresolution with Coupled Backpropagation (SR-CBP) which builds two Coupled Auto-encoder Networks (CAN), resp. the high-resolution...
In this paper an attempt has been made to predict the solar irradiance values for multiple look ahead time predictions with time intervals as small as fifteen minutes. The recurrent neural networks in the past have been implemented on datasets with an interval of at least 30 minutes. The recurrent neural network was trained using backpropagation through time and the prediction was done using only...
We examine a potential technique of performing a classification task based on compressively sensed (CS) data, skipping a computationally expensive reconstruction step. A deep Boltzmann machine is trained on a compressive representation of MNIST handwritten digit data, using a random orthoprojector sensing matrix. The network is first pre-trained on uncompressed data in order to learn the structure...
Vehicle detection can provide volumes of useful data for city planning and transport management. It has always been a challenging task because of various complicated backgrounds and the relatively small sizes of targets, especially in high resolution satellite images. A novel model called joint-layer deep convolutional neural networks (JLDCNNs), which joins features in the higher layers and the lower...
The Artificial Neural Network (ANN) is a branch of science in the field of artificial intelligence and is created from adapting the workings of the human brain. Backpropagation (BP) and Learning Vector Quantisation (LVQ) are two of many methods used to recognise patterns. Both are supervised training methods with different approaches. BP uses an error value to recognise patterns or images, while LVQ...
We introduce a multi-tiered neural network architecture that accurately classifies malignant breast tissue from benign breast tissue. The methodology implemented six different backpropagation neural network (BNN) architectures on 180 malignant and 180 benign breast tissue impedance data files sampled at 47 frequencies from 1 hertz (Hz) to 32 megahertz (MHz). The data were collected utilizing a NovaScan...
Malaria is a serious public health problem in Indonesia. Conventional methods of identification malaria parasite are generally carried out by paramedics when they are thoroughly examine blood performed using a microscope. This way is currently used anywhere, because it is cheap and it has good accuracy than others. However, this conventional methods can make a difference if the diagnosis is made by...
Epilepsy is a central system disorder of human brain in which nerve cell activity becomes disrupted, causing seizures or periods of unsual behaviour, sensations and sometimes loss of consciousness. The electroencephalogram (EEG) is a measure of brain waves can be used in evaluation of brain disorders, one of which epilepsy. In this paper, epilepsy detection system on EEG data is built using combination...
Modeling awareness is an important topic in the computer science as it is closely related to preparing systems that know what is needed (e.g. data accumulated or ignored, effector activated) to achieve a given goal. Preparing tools to build and compare dedicated or general aware computational systems can lead to step-by-step hierarchical construction of intelligent solutions. Within this text we show...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.