ResMLP: Feedforward Networks for Image Classification With Data-Efficient Training.

The learning speed of the feed forward neural network takes a lot of time to be trained which is a major drawback in their applications since the past decades. The key reasons behind may be due to the slow gradient-based learning algorithms which are extensively used to train the neural networks or due to the parameters in the networks which are tuned iteratively using some learning algorithms. Thus, in order to eradicate the above pitfalls, a new learning algorithm was proposed known as Extreme Learning Machines (ELM). This algorithm tries to compute Hidden-layer-output matrix that is made of randomly assigned input layer and hidden layer weights and randomly assigned biases. Unlike the other feedforward networks, ELM has the access of the whole training dataset before going into the computation part. Here, we have devised a new two-layer-feedforward network (TFFN) for ELM in a new manner with randomly assigning the weights and biases in both the hidden layers, which then calculates the output-hidden layer weights using the Moore-Penrose generalized inverse. TFFN doesn't restricts the algorithm to fix the number of hidden neurons that the algorithm should have. Rather it searches the space which gives an optimized result in the neurons combination in both the hidden layers. This algorithm provides a good generalization capability than the parent Extreme Learning Machines at an extremely fast learning speed. Here, we have experimented the algorithm on various types of datasets and various popular algorithm to find the performances and report a comparison.

On computing decision regions with neural nets

Inter-areal information transmission in the brain cortex relates to cognitive functions. Researches used to pay attention to activity pattern transmission, signals gating, or routing in neuronal networks. However, the underlying mechanism of simultaneous transmission of multiple neural signals in the same channel across networks remains unclear. In this work, we construct a two-layer feedforward neuronal network (sender-receiver) with each layer's intrinsic rhythms consisting of slow- (low-frequency) and fast- gamma rhythms (high-frequency), investigating how to realize simultaneous transmission of multiple signals in neuronal systems. With the aid of resonance and frequency analysis, it is shown that low- and high-frequency signals can be transmitted simultaneously in such a feedforward network through frequency division multiplexing (FDM) communication. The transmission performance is related to the local resonance, connectivity, as well as background noise. Moreover, low- and high-frequency signals can also be gated or selected with appropriate adjustments of recurrent connection strength and delay, and background noise. Our model might provide a novel insight into the underlying mechanism of complex signals communication between different cortex areas.

To address the limitations of temporal dependency modeling and the inadequate fusion of long-term information in vessel trajectory prediction, this paper proposes a vessel trajectory prediction model based on an improved Transformer architecture. The proposed model adopts a hybrid architecture that integrates LSTM and Transformer attention mechanisms. Specifically, a single-layer LSTM is first employed to extract sequential features and generate a sequence feature tensor. Subsequently, a standard Transformer self-attention sublayer is introduced to perform global modeling and attention-based weighting on the feature sequence extracted by the LSTM. Residual connections and layer normalization are employed to enhance model stability. Then, a two-layer feed-forward network (FFN), consisting of two Dense layers, is applied to perform nonlinear transformations on the features. The output is again processed with residual connections and normalization, forming a complete Transformer encoder structure. Finally, a GlobalAveragePooling1D layer is applied along the temporal dimension to smooth the sequence, and a Dense layer outputs the next-step predictions for longitude (lon), latitude (lat), speed over ground (sog), and course over ground (cog). To validate the effectiveness of the proposed model, comparative experiments are conducted against conventional GRU, LSTM, and the vanilla Transformer models. Experimental results demonstrate that the proposed approach outperforms the baselines across multiple evaluation metrics, exhibiting superior trajectory-modeling capability and prediction accuracy, and thereby significantly enhancing vessel trajectory forecasting performance.

Presents experimental results which show that feedforward neural networks are suitable for analog IC fault diagnosis. The results suggest that feedforward networks provide a cost-efficient method for IC fault diagnosis in large-scale production. The authors compare the diagnostic accuracy and the computational requirements of a simple feedforward network against that of Gaussian maximum likelihood and K-nearest neighbors classifiers. The feedforward network was found to provide an order-of-magnitude improvement in diagnostic speed while consistently performing as well as or better than any of the other classifiers in terms of accuracy.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Angiotensin I converting enzyme (ACE) gene, as a component of Renin-Angiotensin System (RAS), regulates blood pressure as it converts somatic isozyme Angiotensin I into physiologically active peptide Angiotensin II and simultaneously brakes down bradykinins. Over 100 polymorphisms are reported for ACE gene. Most of these polymorphisms having no phenotypic effect relay the attention towards polymorphisms based on insertions (I) or deletions (D) of a 287 bp Alu repeat sequence in 16th intron. There are three possible genotypes for the stated polymorphism: DD, DI or II. There have been a lot of studies searching for direct associations between ACE polymorphisms and performance phenotypes along different sports requiring power or endurance. The previous experiments are based on the performance criteria but direct associations of ACE polymorphisms are not fully understood until today. We believe different approaches may aid scientist to plot the big picture. A sample population of 101 individuals from Bosnia and Herzegovina contributed to sample pool of the initial project. Buccal swabs from 101 samples were collected along with a phenotypic structure and environmental characteristic survey which was filled by each individual himself/herself. Genotypes of the individuals were obtained after isolation, amplification and gel electrophoresis of biological samples collected as buccal swabs. A total of 165 artificial neural network (ANN) models were developed considering the input parameters, possible genotype outputs, applied algorithm and sample size. The aim of developing various ANNs was to validate a possible ACE polymorphism genotype prediction algorithm based on phenotypic and environmental characteristics of individuals, in other words, without any biological testing. A two-layered feed-forward network, with sigmoid hidden neurons was designed to perform the classification of input data. Trainscg (Scaled Conjugate Gradient) activation function was used in hidden layer since classification of data was non-linear. All ANN models were trained with scaled conjugate gradient backpropagation. ANN models differing in the parameters has shown different accuracy in the results. Most outstanding result was observed in the ANN build composed of 2 distinctive layers with 500 neurons in the first and 3 neurons in the second layer. Trained with 70% of samples and verified with 15% of samples and validated with an additional 50 samples. Training set was composed of the following subject parameters; gender, eye color, hair color, height, weight, presence of hypertension in family and presence of cardiovascular diseases in family. The highest prediction accuracy was obtained as 86,6 % training score, 78,6 % testing score and 80,2 % overall score in genotype prediction for ACE polymorphisms. With further development of data collection and high resolution analysis, overall score could be boosted. Also, phenotypical data can be applied as markers of genotypes in ACE polymorphisms.

PurposeThe purpose of this paper is to predict the machining performance of electrical discharge machining of Ti-13Nb-13Zr (TNZ) alloy, a promising biomedical alloy, using artificial neural networks (ANN) models.Design/methodology/approachIn the research, three major performance characteristics, i.e. the material removal rate (MRR), tool wear rate (TWR) and surface roughness (SR), were chosen for the study. The input parameters for machining were the voltage, current, pulse-on time and pulse-off time. For the ANN model, a two-layer feedforward network with sigmoid hidden neurons and linear output neurons were chosen. Levenberg–Marquardt backpropagation algorithm was used to train the neural networks.FindingsThe optimal ANN structure comprises four neurons in input layer, ten neurons in hidden layer and one neuron in the output layer (4–10-1). In predicting MRR, the 60–20-20 data split provides the lowest MSE (0.0021179) and highest R-value for training (0.99976). On the contrary, the 70–15-15 data split results in the best performance in predicting both TWR and SR. The model achieves the lowest MSE and highest R-value for training in predicting TWR as 1.17E-06 and 0.84488, respectively. Increasing the number of hidden neurons of the network further deteriorates the performance. In predicting SR, the authors find the best MSE and R-value as 0.86748 and 0.94024, respectively.Originality/valueThis is a novel approach in performance prediction of electrical discharge machining in terms of new workpiece material (TNZ alloys).

This paper presents development of Artificial Neural Network (ANN) for prediction of the size of nanoparticles (NP) and microspore surface area (MSA). Developed neural network architecture has the following three inputs: the concentration of the biodegradable polymer in the organic phase, surfactant concentration in the aqueous phase and the homogenizing pressure. Two-layer feedforward network with a sigmoid transfer function in the hidden layer and a linear transfer function in the output layer is trained, using Levenberg-Marquardt training algorithm. For training of this network, as well as for subsequent validation, 36 samples were used. From 36 samples which were used for subsequent validation in this ANN, 80,5% of them had highest accuracy while 19,5% of output data had insignificant differences comparing to experimental values.

CFD data based neural network functions for predicting hydrodynamic performance of a low-pitch marine cycloidal propeller

The article deals with the formation of adhesion bonds in the process of applying coatings to machine building products by the method of cladding by a flexible tool. The main points of the mathematical model for determining the adhesion of applied coatings are shown. The calculation in the model is based on the use of the coefficient of relative strength for the particles adhesion of the coating material to the workpiece, the contact temperature of the particles with the base material during their interaction, and the energy of thermal activation of the surface as a result of the chemical reaction considering the energy of mechanical activation. The dataset of the coefficient of relative strength for the adhesion is given for varying the circumferential velocity of the tool, the diameter, and the length of the wires from which the flexible tool is assembled. The modeling of the dependences of the relative strength coefficient is considered as an approximation problem using artificial neural networks. The input data are the technological and structural parameters of the cladding. For modeling, a two-layer feed-forward network is used. For a problem of partitioning input parameters into regions of permissible or unacceptable from the point of view of providing a given coefficient of relative strength, a similar neural network with an output layer with a softmax output neuron is used. The use of two approaches makes it possible to accelerate the design of the cladding operation while increasing the reliability of the decisions made.

In this paper development of Artificial Neural Network for classification of prediabetes and type 2 diabetes (T2D) is presented. For development of this system 310 samples consisting of information about Fasting Plasma Glucose (FPG) and blood test called HbA1c were used. All samples were obtained from several healthcare institutions in Bosnia and Herzegovina, and diagnosis of prediabetes, T2D and healthy patients in this dataset were established by medical professionals. Two-layer feedforward backpropagation network with 15 neurons in hidden layer and sigmoid transfer function, used for classification of prediabetes and T2D in this paper, was trained with 190 samples.

Ultrasound (US) imaging is the initial phase in the preliminary diagnosis for the treatment of kidney diseases, particularly to estimate kidney size, shape and position, to give information about kidney function, and to help in diagnosis of abnormalities like cysts, stones, junctional parenchyma and tumors which is shown in Figs. 7–9. This study proposes Grey Level Co-occurrence Matrix (GLCM)-based Probabilistic Principal Component Analysis (PPCA) and Artificial Neural Network (ANN) method for the classification of kidney images. Grey Wolf Optimization (GWO) is used to update the current positions of abnormal kidney images in the discrete searching space, thus getting the optimal feature subset for better classification purposes based on Feed Forward Neural Network (FFNN). The scanned image is pre-processed and the required features are extracted by GLCM, among those, some features are selected by PPCA. Feed Forward Back propagation Neural Network (FFBN) is used to classify the normalities and abnormalities in the part of kidney images. The proposed methodology is implemented in MATLAB platform and the analyzed result produces 98% accuracy using GWO-FFBN technique.

In this paper, we showed a method to forecast the stock price using neural networks. Predicting the stock market is very difficult since it depends on several known and unknown factors. In recent years, one of the techniques that have been used popularly in this area is artificial neural network. The power of neural network is its ability to model a nonlinear process without a priori knowledge about the nature of the process. We used both feed forward neural network and simple recurrent neural network, trained by time and profit based back propagation algorithm with early stopping to make the prediction. The integration of profit and time factors with training procedure made an improvement in forecasted results for feed forward neural network. Moreover, the simple recurrent neural network with its ` time capture' capabilities had better forecasted results than feed forward neural network in all experiments.

This paper presents an algorithm to detect Parkinson’s disease using voiced features and an artificial neural network (ANN). It uses 44 audio features extracted from the sustained vowel’/a/’ sound of the Parkinson’s disease patients and healthy subjects. To fully characterize a healthy and a Parkinson’s patient’s voice, audio features of different domains have been investigated. A simple two-layer feed-forward neural network (FFNN) has been deployed to avoid an overwhelming computational burden on the system. Significant statistical parameters are measured to evaluate the performance of the proposed algorithm. The simulation results show that the proposed algorithm achieves an accuracy of 85% and G-mean of 85.19% in identifying Parkinson’s patients.

Rainfall forecasting is very challenging due to its uncertain nature and dynamic climate change. It's always been a challenging task for meteorologists. In various papers for rainfall prediction, different Data Mining and Machine Learning (ML) techniques have been used. These techniques show better predictive accuracy. A deep learning approach has been used in this study to analyze the rainfall data of the Karnataka Subdivision. Three deep learning methods have been used for prediction such as Artificial Neural Network (ANN) - Feed Forward Neural Network, Simple Recurrent Neural Network (RNN), and the Long Short-Term Memory (LSTM) optimized RNN Technique. In this paper, a comparative study of these three techniques for monthly rainfall prediction has been given and the prediction performance of these three techniques has been evaluated using the Mean Absolute Percentage Error (MAPE%) and a Root Mean Squared Error (RMSE%). The results show that the LSTM Model shows better performance as compared to ANN and RNN for Prediction. The LSTM model shows better performance with mini-mum Mean Absolute Percentage Error (MAPE%) and Root Mean Squared Error (RMSE%).