Input Weight Vector Research Articles

Hybrid Convolutional Neural Network with PSO Based Severe Dengue Prognosis Method in Human Genome Data

Dengue is one of the most significant diseases transmitted by arthropods in the world. Dengue phenotypes are focused on documented inaccuracies in the laboratory and clinical studies. In countries with a high incidence of this disease, early diagnosis of dengue is still a concern for public health. Deep learning has been developed as a highly versatile and accurate methodology for classification and regression, which requires small adjustment, interpretable results, and the prediction of risk for complex diseases. This work is motivated by the inclusion of the Particle Swarm Optimization (PSO) algorithm for the fine-tuning of the model's parameters in the convolutional neural network (CNN). The use of this PSO was used to forecast patients with extreme dengue, and to refine the input weight vector and CNN parameters to achieve anticipated precision, and to prevent premature convergence towards local optimum conditions.

Simple Modification of Oja Rule Limits L1-Norm of Weight Vector and Leads to Sparse Connectivity

This letter describes a simple modification of the Oja learning rule, which asymptotically constrains the L1-norm of an input weight vector instead of the L2-norm as in the original rule. This constraining is local as opposed to commonly used instant normalizations, which require the knowledge of all input weights of a neuron to update each one of them individually. The proposed rule converges to a weight vector that is sparser (has more zero weights) than the vector learned by the original Oja rule with or without the zero bound, which could explain the developmental synaptic pruning.

Modelling procedures for directed network of data blocks

Here are presented procedures for modelling data in a network. The methods are extensions of PCA or PLS regression to a forward network of data blocks. It is assumed that the data blocks are organised in a network such that one data block leads to one or more other data blocks. The procedures are stepwise ones. At each step a passage through the network is carried out. From the input weight vectors of the input or starting blocks, the score and loading vectors of all data blocks are computed. It is investigated if some score/loading vectors are not significant. If some are, they are deleted and revised estimation of the input weights are carried out. When one step is finished, all data matrices are adjusted for score and loading vectors found. A new passage through the network is carried out on the reduced matrices. If no significant loading/score vectors are found for a given set of input weights, the modelling stops. In case of one data block, the algorithm reduces to PCA. In case of two data blocks it reduces to PLS regression. Most methods used in PCA or PLS regression can be applied to this procedure, e.g., cross-validation and re-sampling procedures. It is pointed out, how these methods can be used to extend other regression methods than PCA and PLS regression to a network regression.

Combined Statistical and Multiscale View on Ultrasonic Liver Images for Characterization

Combined Statistical and Multiscale View on Ultrasonic Liver Images for Characterization

An Adaptive Structure Neural Networks with Application to EEG Automatic Seizure Detection

This paper introduces a new algorithm for adaptively adjusting the structure of a multi-layer back-propagation network. The proposed algorithm belongs to the class of neuron generating strategies as opposed to the class of neuron pruning strategies. Initially a “small” multi-layer perceptron network is selected. The stabilized error is used as an index to determine whether the network needs to generate a new neuron or not. If after a period of learning the error is stabilized, but the error is larger than a desired value, then new neuron(s) is (are) generated. The new neurons are placed at locations that contribute most to the network error behavior through the fluctuation in their input weight vectors. Among the features of the new architecture are its improved performance and generalization capabilities compared to a standard fixed-structure back-propagation network. Application to an electroencephalogram (EEG) automatic epileptic seizure detection is presented to illustrate advantages and capabilities of the proposed algorithm. Using an actual data from five patients it is shown that the proposed approach correctly identifies all true seizures that are also identified by an expert physician. The new algorithm provides a reduction of 60–70% in the training epochs as compared to a back-propagation algorithm. Furthermore, it is shown that by utilizing a new training algorithm it is possible to reduce the false seizure detections to zero while resulting in a 5.1% error in identifying the true seizures. Copyright © 1996 Elsevier Science Ltd

Extraction of fuzzy rules using neural networks with structure level adaptation--verification to the diagnosis of hepatobiliary disorders.

This paper presents the reasoning and learning method for fuzzy rules using structure level adaptation of neural networks. In a usual neural network mechanism, we can observe some behaviors during the learning process. Based on such behaviors of neuron activity, we can generate or annihilate the specified neurons respectively in hidden layer to achieve an overall good system. In the method that we have proposed, we have used a procedure to derive the neuron generation/annihilation automatically, and applied such a procedure to the learning system where the experimental data related to hepatobiliary disorders were used. After learning by using randomly chosen data, the proposed system correctly diagnosed over 70% of cases. According to these results, we can find that fuzzy rules have some relationship with the degree of the input weight vector. As a result, we can assume that fuzzy rules for hepatobiliary disorders are extracted from this learned network.