How to optimize neuroscience data utilization and experiment design for advancing brain models of visual and linguistic cognition?

Introduction River flow forecasting has been one of the important challenges of water resources management in recent decades, so many researchers have proposed different methods to improve the performance of forecasting models. In the last decade, artificial intelligence methods have been most widely used in the simulation of various processes, including hydrological processes, due to their flexibility and high accuracy in modeling. However, the results of these methods have always been associated with uncertainty due to several factors such as structure, algorithm, input data, and the type of method chosen for data calibration. One of the methods that can somewhat solve this problem is the uncertainty analysis of the predictions made by these models. Materials and Methods In this study, the uncertainty of the results of artificial neural network (ANN) and support vector machine (SVM) models in predicting the monthly flow of the river has been evaluated. In this research, the time series of the monthly flow of the Ghezelozan River using the data of the Bianlu-Yasaul Stream gauging station in 39 years from 1976 to 2014 was used, where 75% and 25% of the data was used for training and testing the models, respectively. In these models, to estimate the monthly flow of the Ghezelozan River, six different input combinations including the flow of one, two, and three months before and the number of months of the flow were used. Then, the accuracy and performance of the models were compared using the coefficient of determination (R) and root mean square of errors (RMSE). Finally, the uncertainty of the results of ANN and SVM models in predicting the monthly flow of the river was evaluated by the Monte-Carlo method using dfactor and 95PPU values. Results and Discussion The evaluation of the performance of the ANN model shows that the best performance is related to the combination where the flow of the previous two months and the number of the month of the flow are the inputs of the model so that R and RMSE indexes were obtained as 0.757 and 9.45, respectively. In the SVM model for the monthly river flow series, the best performance is related to the combination where the flow of one, two, and three months ago and the number of months of the flow were the inputs of the model, and the R and RMSE indexes for this input pattern were 0.729 and 10.946, respectively. After studying the uncertainty of the models, the results showed that the ANN model has more uncertainty in the output values compared to the SVM model, and this is while the d-factor of the ANN model, both in the training and test phase, it was more than the SVM model. The comparison of the uncertainty analysis of the results of the ANN and SVM models showed that the SVM model with d-factor and 95PPU values equal to 0.155 and 17.241, respectively, compared to the ANN model with d-factor and 95PPU values equal to 0.993 and 85.470, respectively, has less uncertainty in the output values. So the number of observation data placed in the 95% confidence range (95PPU) of the ANN model compared to the SVM model has increased significantly in both the training and testing phases. Also, the results showed that both models have more uncertainty in the months with a large volume of water, which can be due to the complexity of the process and the involvement of uncertain factors in these months, as well as the effect of factors that are not considered in the structure of predictive models. Conclusion The results of ANN and SVM models in predicting the monthly flow of the Ghezelozan River showed that although the ANN model with R-value equal to 0.757 and RMSE value equal to 9.45 has a good performance compared to the SVM model with R-value equal to 0.729 and RMSE value equal to 10.946 in predicting the river flow, the results of this model are associated with high uncertainty. The comparison of the uncertainty analysis of the results of ANN and SVM models by Monte-Carlo method showed that the SVM model with dfactor and 95PPU values equal to 0.155 and 17.241, respectively, compared to the ANN model with d-factor and 95PPU values equal to 0.993 and 85.470, respectively, has less uncertainty in predicting the monthly flow of the Ghezelozan River and it is better than ANN model. According to the results of this research, taking into account the fact that advanced artificial intelligence models also have uncertainty, it is necessary to apply these methods in the fields of risk management and future planning to obtain the best performance.

Kernel-based LFP estimation in detailed large-scale spiking network model of mouse visual cortex.

This article compares two modeling approaches for optimal operation of a turbo chiller installed in an office building: (1) a machine learning model developed with artificial neural network (ANN) and (2) a hybrid machine learning model developed with the ANN model and available physical knowledge of the chiller. Before developing the ANN model of the chiller, the authors used Gaussian mixture model in order to check the validity of measured data. Then, the hybrid model was developed by combining the ANN model and physics-based regression equations from the EnergyPlus engineering reference. It was found that both the ANN and hybrid ANN model are satisfactory to predict the chiller’s power consumption: mean bias error (MBE) = −2.63%, coefficient of variation of the root mean square error (CVRMSE) = 8.05% by the ANN model; MBE = −3.99%, CVRMSE = 11.98% by the hybrid ANN model. However, the hybrid model requires fewer inputs (four inputs) than the ANN model (eight inputs). The energy savings of both models are similar coefficient of performance (COP) = 4.32 by the optimal operation of the ANN model; COP = 4.44 by the optimal operation of the hybrid ANN model. In addition, the hybrid ANN model can be applied where the ANN model is unable to provide accurate predictions.

Neural network techniques have been widely adopted for various tasks, including sequence time-series prediction. Researchers tend to develop large-scale neural network models with multi-level frameworks and extensive numbers of parameters. However, this strategy might not be appropriate for the dynamic response prediction (DRP) problem because large-scale neural network models may tend to overfit the data rather than learn the system’s inherent dynamic property. This paper advocates that neural network models with concise architectures and the ability to make step-by-step moving-forward predictions are better suited for DRP problems. Building on this opinion, the Elman recurrent, long short-term memory, discrete-time state-space, and autoregressive neural networks are adopted in this paper to construct surrogate models for DRP tasks. Numerical and experimental evaluations of these models and other large-scale neural network models are conducted, demonstrating that the four selected types of neural network models, built with limited numbers of parameters, could achieve good performance and minimal overfitting, with overall R-squared values exceeding 90%. It is recommended to utilize these small-scale neural networks or their modifications for DRP problems rather than to construct complicated and large-scale neural networks.

Dendritic synaptic integration in central neurons

The search for better climate change adaptation techniques for addressing environmental and economic issues due to changing climate is of paramount interest in the current era. One of the many ways Pacific Island regions and its people get affected is by dry spells and drought events from extreme climates. A drought is simply a prolonged shortage of water supply in an area. The impact of drought varies both temporally and spatially that can be catastrophic for such regions with lack of resources and facilities to mitigate the drought impacts. Therefore, forecasting drought events using predictive models that have practical implications for understanding drought hydrology and water resources management can allow enough time to take appropriate adaption measures. This study investigates the feasibility of the Artificial Neural Network (ANN) algorithms for prediction of a drought index: Standardized Precipitation-Evapotranspiration Index (SPEI). The purpose of the study was to develop an ANN model to predict the index in two selected regions in Queensland, Australia. The first region, is named as the grassland and the second as the temperate region. The monthly gridded meteorological variables (precipitation, maximum and minimum temperature) that acted as input parameters in ANN model were obtained from Australian Water Availability Project (AWAP) for 1915–2013 period. The potential evapotranspiration (PET), calculated using thornthwaite method, was also an input variable, while SPEI was the predictand for the ANN model. The input data were divided into training (80%), validation (10%) and testing (10%) sets. To determine the optimum ANN model, the Levenberg-Marquardt and Broyden-Fletcher-Goldfarb-Shanno quasi-Newton backpropagation algorithms were used for training the ANN network and the tangent sigmoid, logarithmic sigmoid and linear activation algorithms were used for hidden transfer and output functions. The best architecture of input-hidden neuron-output neurons was 4-28-1 and 4-27-1 for grassland and temperate region, respectively. For evaluation and selection of the optimum ANN model, the statistical metrics: Coefficient of Determination (R 2 ), Willmott’s Index of Agreement (d), Nash-Sutcliffe Coefficient of Efficiency (E), Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) were employed. The R 2 , d, E, RMSE and MAE for optimum ANN models were 0.9839, 0.9909, 0.9838, 0.1338, 0.0882 and 0.9886, 0.9935, 0.9874, 0.1198, 0.0814 for grassland and temperate region, respectively. When prediction errors were analysed, a value of 0.0025 to 0.8224 was obtained for the grassland region, and a value of 0.0113 to 0.6667 was obtained for the temperate region, indicating that the ANN model exhibit a good skill in predicting the monthly SPEI. Based on the evaluation and statistical analysis of the predicted SPEI and its errors in the test period, we conclude that the ANN model can be used as a useful data-driven tool for forecasting drought events. Broadly, the ANN model can be applied for prediction of other climate related variables, and therefore can play a vital role in the development of climate change adaptation and mitigation plans in developed and developing nations, and most importantly, in the Pacific Island Nations where drought events have a detrimental impact on economic development.

Chapter 4 - Concepts, procedures, and applications of artificial neural network models in streamflow forecasting

Artificial neural network algorithms to predict the bond strength of reinforced concrete: Coupled effect of corrosion, concrete cover, and compressive strength

Temperature (T) and moisture content (MC) of non-homogenous food undergoing microwave-vacuum (MV) drying (MVD) are directly dependent on microwave power, vacuum pressure, and the product's physical properties. A two-hidden-layer Artificial Neural Network (ANN) model was developed in an earlier study to predict temperature and moisture content of the product at a given time based on the present state of product conditions and process control parameters. This approach either provided lowest error in temperature prediction or in moisture content prediction but not the lowest error in both the prediction parameters simultaneously. The main objective of this work was to improve the performance of the ANN model for temperature and moisture content predictions in MV dried samples. Experimental data obtained from MVD of tomato slices at different drying conditions was normalized and divided into two groups for training and validating. The parallel dynamic ANN model consisted of two double-hidden-layer feed-forward ANN models with varying node numbers (10, 20, and 30). These models were separately trained, simultaneously for moisture content as well as temperature, with the Levenberg-Marquardt algorithm. Inputs for the ANN models were magnetron on-off status, vacuum pressure, temperature, and moisture content at time `ti'. The previous temperature and moisture content data at time `ti-1, i-2, , i-n' where n = 0, 10, 20, and 30 were also added to the input layer. Outputs from the ANN models were temperature and moisture content at time `ti+1'. The results indicated that the dynamic ANN model working in parallel with the previous temperature and moisture content data provided results that are more accurate and required less training time than those of ordinary ANN models. Model simulation may supply essential information regarding temperature and moisture content of non-homogenous foods corresponding to microwave power and vacuum pressure levels to the predictive control system. Therefore, improved drying efficiencies and thermal damage prevention may be achieved.

The forward osmosis (FO) process is an emerging technology that has been considered as an alternative to desalination due to its low energy consumption and less severe reversible fouling. Artificial neural networks (ANNs) and response surface methodology (RSM) have become popular for the modeling and optimization of membrane processes. RSM requires the data on a specific experimental design whereas ANN does not. In this work, a combined ANN-RSM approach is presented to predict and optimize the membrane flux for the FO process. The ANN model, developed based on an experimental study, is used to predict the membrane flux for the experimental design in order to create the RSM model for optimization. A Box–Behnken design (BBD) is used to develop a response surface design where the ANN model evaluates the responses. The input variables were osmotic pressure difference, feed solution (FS) velocity, draw solution (DS) velocity, FS temperature, and DS temperature. The R2 obtained for the developed ANN and RSM model are 0.98036 and 0.9408, respectively. The weights of the ANN model and the response surface plots were used to optimize and study the influence of the operating conditions on the membrane flux.

The bi-reforming of methane (BRM) is a promising process which converts greenhouse gases to syngas with a flexible H2/CO ratio. As there are many factors that affect this process, the coupled effects of multi-parameters on the BRM product are investigated based on Gibbs free energy minimization. Establishing a reliable model is the foundation of process optimization. When three input parameters are changed simultaneously, the resulting BRM products are used as the dataset to train three artificial neural network (ANN) models, which aim to establish the BRM prediction model. Finally, the trained ANN models are used to predict the BRM products when the conditions vary in and beyond the training range to test their performances. Results show that increasing temperature is beneficial to the conversion of CH4. When the molar flow of H2O is at a low level, the increase in CO2 can enhance the H2 generation. While it is more than 0.200 kmol/h, increasing the CO2 flowrate leads to the increase and then decrease in the H2 molar flow in the reforming products. When the numbers of hidden layer neurons in ANN models are set as (3, 3), (3, 6) and (6, 6), all the correlation coefficients of training, validation and test are higher than 0.995. When these ANN models are used to predict the BRM products, the variation range of the prediction error becomes narrower, and the standard deviation decreases with the increase in neuron number. This demonstrates that the ANN model with more neurons has a higher accuracy. The ANN model with neuron numbers of (6, 6) can be used to predict the BRM products even when the operating conditions are beyond the training ranges, demonstrating that this model has good extension performance. This work lays the foundation for an artificial intelligent model for the BRM process, and established ANN models can be further used to optimize the operating parameters in future work.

Groundwater level fluctuation modeling is a prime need for effective utilization and planning the conjunctive use in any basin.The application of Artificial Neural Network (ANN) and hybrid Wavelet ANN (WANN) models was investigated in predicting Groundwater level fluctuations. The RMSE of ANN model during calibration and validation were found to be 0.2868 and 0.3648 respectively, whereas for the WANN model the respective values were 0.1946 and 0.1695. Efficiencies during calibration and validation for ANN model were 0.8862 per cent and 0.8465 per cent respectively, whereas for WANN model were found to be much higher with the respective values of 0.9436 per cent and 0.9568 per cent indicating substantial improvement in the model performance. Hence hybrid ANN model is the promising tool to predict water table fluctuation as compared to ANN model.

This study deals with the development of an artificial neural network (ANN) and a multiple regression (MR) model that can be employed for estimating the California bearing ratio (CBR) value of some Aegean sands. To achieve this, the results of CBR tests performed on the compacted specimens of nine different Aegean sands with varying soil properties were used in the development of the ANN and MR models. The results of the ANN and MR models were compared with those obtained from the experiments. It is found that the CBR values predicted from the ANN model matched the experimental values much better than the MR model. Moreover, several performance indices, such as coefficient of determination, root-mean-square error, mean absolute error, and variance, were used to evaluate the prediction performance of the ANN and MR models. The ANN model has shown higher prediction performance than the MR model based on the performance indices, which demonstrates the usefulness and efficiency of the ANN model. Thus, the ANN model can be used to predict CBR value of the Aegean sands included in this study as an inexpensive substitute for the laboratory testing, quite easily and efficiently.

Most reports compare artificial neural network (ANN) models and logistic regression models in only a single data set, and the essential issue of internal validity (reproducibility) of the models has not been adequately addressed. This study proposes to validate the use of the ANN model for predicting in-hospital mortality after traumatic brain injury (TBI) surgery and to compare the predictive accuracy of ANN with that of the logistic regression model. The authors of this study retrospectively analyzed 16,956 patients with TBI nationwide who were surgically treated in Taiwan between 1998 and 2009. For every 1000 pairs of ANN and logistic regression models, the area under the receiver operating characteristic curve (AUC), Hosmer-Lemeshow statistics, and accuracy rate were calculated and compared using paired t-tests. A global sensitivity analysis was also performed to assess the relative importance of input parameters in the ANN model and to rank the variables in order of importance. The ANN model outperformed the logistic regression model in terms of accuracy in 95.15% of cases, in terms of Hosmer-Lemeshow statistics in 43.68% of cases, and in terms of the AUC in 89.14% of cases. The global sensitivity analysis of in-hospital mortality also showed that the most influential (sensitive) parameters in the ANN model were surgeon volume followed by hospital volume, Charlson comorbidity index score, length of stay, sex, and age. This work supports the continued use of ANNs for predictive modeling of neurosurgery outcomes. However, further studies are needed to confirm the clinical efficacy of the proposed model.

In this study, artificial neural network (ANN) and thermodynamic models were developed for prediction of the heat capacity (C P ) of amine-based solvents. For ANN model, independent variables such as concentration, temperature, molecular weight and CO2 loading of amine were selected as the inputs of the model. The significance of the input variables of the ANN model on the C P values was investigated statistically by analyzing of correlation matrix. A thermodynamic model based on the Redlich-Kister equation was used to correlate the excess molar heat capacity $$ \left({C}_P^E\right) $$ data as function of temperature. In addition, the effects of temperature and CO2 loading at different concentrations of conventional amines on the C P values were investigated. Both models were validated against experimental data and very good results were obtained between two mentioned models and experimental data of C P collected from various literatures. The AARD between ANN model results and experimental data of C P for 47 systems of amine-based solvents studied was 4.3%. For conventional amines, the AARD for ANN model and thermodynamic model in comparison with experimental data were 0.59% and 0.57%, respectively. The results showed that both ANN and Redlich-Kister models can be used as a practical tool for simulation and designing of CO2 removal processes by using amine solutions.