This work introduces cutting-edge research to predict heart diseases. This application designed for accurate and personalized assessment of cardiovascular health. This work aims to provide an overall understanding of ...
详细信息
Artificial neural networks(ANNs) have been widely used to fit the potential energy surface. Levenberg-Maquard(LM) algorithm is a powerful neural network training algorithm, recognized as one of the fastest ways to tra...
详细信息
Artificial neural networks(ANNs) have been widely used to fit the potential energy surface. Levenberg-Maquard(LM) algorithm is a powerful neural network training algorithm, recognized as one of the fastest ways to train ANNs of medium size. However, its efficient is still insufficient for a very large system. extremelearningmachine is a recently proposed algorithm which resolves the we ights in one shot by a linear equation and is thus extremely fast. However its accuracy remains a bottleneck because of its random nature. This work attempts to take advantages of both algorithms. This so-called LM-ELM algorithm performs one ELM halfway forward and one LM halfway backward iteratively, thus reduces the time-consuming LM steps.[1] The LM-ELM algorithm has been tested using the PES of H+H2, CO2+Ni(100), and CH4+Ni(111) systems, which performs better in the convergence speed and generalization ability comparing to the sole LM algorithm.[2]
Atomistic dimensional neural networks(ANNs) have been widely used to fit the potential energy surface. ANNs can be used for fitting a very large system including thousands of atoms by designing a different neural netw...
详细信息
Atomistic dimensional neural networks(ANNs) have been widely used to fit the potential energy surface. ANNs can be used for fitting a very large system including thousands of atoms by designing a different neural networks for different atom with distinct element. Levenberg-Maquardt(LM) algorithm is a powerful neural network training algorithm. However, its efficient is still insufficient for a very large system. extremelearningmachine is a recently proposed algorithm which resolves the weights in one shot by a linear equation and is thus extremely fast. However its accuracy remains a bottleneck because of its random nature. This work attempts to take advantages of both algorithms and constructs the ANNs with a modified version of Behler and Parrinello symmetry functions[1]. This so-called LM-ELM algorithm performs one ELM halfway forward and one LM halfway backward iteratively, thus reduces the time-consuming LM steps.[2] The LM-ELM algorithm has been tested using the PES of H+H2 and H2 CC systems, which performs better in the convergence speed and generalization ability comparing to the sole LM algorithm.[3]
Uniaxial compressive strength (UCS) is substantially used mechanical parameters to observe and classification of rocks, but this test is subsersive, taking a long time and required well equipped laboratory conditions ...
详细信息
Uniaxial compressive strength (UCS) is substantially used mechanical parameters to observe and classification of rocks, but this test is subsersive, taking a long time and required well equipped laboratory conditions and properly prepared samples. Therefore it is important to estimate this parameter from other physico-mechanical rock parameters that are nondestructive, easy to prepare samples and required less time. machinelearning methods which are among these methods and increase their importance and validty are Multilayer Perceptron Neural Network (MLPNN), M5 Model Tree (M5MT), extremelearningmachine (ELM) methods. In this study, Brazilian tensile strength, ultrasonic P-wave velocity, shore hardness tests of different rock types (Basalt, limestone, dolostone) were performed. The results were used for estimating UCS using MLPNN, M5MT, ELM methods. The validation of models were checked root mean squared error (RMSE), mean absolute error (MAE), variance account for (VAF) and coefficient of determination (R-2) and a10-index. Weights and bias values for MLPNN and ELM approaches and the tree structure for the M5MT method are presented. The result indicated MLPNN model outperforms the other models. Based on the result of predictive models with RMSE, MAE, VAF and R-2 equal to RMSE: 1.3421, MAE: 0.7985, VAF: 99.7409, R-2: 0.9982%.
In order to classify data and improve extremelearningmachine (ELM), this study explains how a hybrid optimization-driven ELM technique was devised. Input data are pre-processed in order to compute missing values and...
详细信息
In order to classify data and improve extremelearningmachine (ELM), this study explains how a hybrid optimization-driven ELM technique was devised. Input data are pre-processed in order to compute missing values and convert data to numerical values using the exponential kernel transform. The Jaro-Winkler distance is used to identify the relevant features. The feed-forward neural network classifier is used to categorize the data, and it is trained using a hybrid optimization technique called the modified enhanced Invasive Weed, a meta heuristic algorithm, and Cuckoo Search, a non-linear optimization algorithm ELM. The enhanced Invasive Weed optimization (IWO) algorithm and the enhanced Cuckoo Search (CS) algorithm are combined to create the modified CSIWO. The experimental findings presented in this work demonstrate the viability and efficacy of the created ELM method based on CSIWO, with good experimental result as compared to other ELM techniques.
暂无评论