Marginal maximum likelihood (MML) estimation is the preferred approach to fitting item response theory models in psychometrics due to the MML estimator’s consistency, normality, and efficiency as the sample size tends to infinity. However, state-of-the-art MML estimation procedures such as the Metropolis–Hastings Robbins–Monro (MH-RM) algorithm as well as approximate MML estimation procedures such as variational inference (VI) are computationally time-consuming when the sample size and the number of latent factors are very large. In this work, we investigate a deep learning-based VI algorithm for exploratory item factor analysis (IFA) that is computationally fast even in large data sets with many latent factors. The proposed approach applies a deep artificial neural network model called an importance-weighted autoencoder (IWAE) for exploratory IFA. The IWAE approximates the MML estimator using an importance sampling technique wherein increasing the number of importance-weighted (IW) samples drawn during fitting improves the approximation, typically at the cost of decreased computational efficiency. We provide a real data application that recovers results aligning with psychological theory across random starts. Via simulation studies, we show that the IWAE yields more accurate estimates as either the sample size or the number of IW samples increases (although factor correlation and intercepts estimates exhibit some bias) and obtains similar results to MH-RM in less time. Our simulations also suggest that the proposed approach performs similarly to and is potentially faster than constrained joint maximum likelihood estimation, a fast procedure that is consistent when the sample size and the number of items simultaneously tend to infinity.

Transfer learning has been highlighted as a promising framework to increase the accuracy of the data-driven model in the case of data sparsity, specifically by leveraging pretrained knowledge to the training of the target model. The objective of this study is to evaluate whether the number of requisite training samples can be reduced with the use of various transfer learning models for predicting, for example, the chemical source terms of the data-driven reduced-order modeling (ROM) that represents the homogeneous ignition of a hydrogen/air mixture. Principal component analysis is applied to reduce the dimensionality of the hydrogen/air mixture in composition space. Artificial neural networks (ANNs) are used to regress the reaction rates of principal components, and subsequently, a system of ordinary differential equations is solved. As the number of training samples decreases in the target task, the ROM fails to predict the ignition evolution of a hydrogen/air mixture. Three transfer learning strategies are then applied to the training of the ANN model with a sparse dataset. The performance of the ROM with a sparse dataset is remarkably enhanced if the training of the ANN model is restricted by a regularization term that controls the degree of knowledge transfer from source to target tasks. To this end, a novel transfer learning method is introduced, Parameter control via Partial Initialization and Regularization (PaPIR), whereby the amount of knowledge transferred is systemically adjusted in terms of the initialization and regularization schemes of the ANN model in the target task.

Soft robotics is rapidly advancing, particularly in medical device applications. A particular miniaturized manipulator design that offers high dexterity, multiple degrees-of-freedom, and better lateral force rendering than competing designs, has great potential for minimally invasive surgery. However, it faces challenges such as the tendency to suddenly and unpredictably deviate in bending plane orientation at higher pressures. In this work, we identified the cause of this deviation as the buckling of the partition wall and proposed design alternatives along with their manufacturing process to address the problem without compromising the original design features. In both simulation and experiment, the novel design managed to achieve a better bending performance in terms of stiffness and reduced deviation of the bending plane. We also developed an artificial neural network-based inverse kinematics model to further improve the performance of the prototype during vectorization. This approach yielded mean absolute errors in orientation of the bending plane below $5^{\circ }$.

Multicomponent systems are representative of the most common real situations as many industrial discharges contain a mixture of several pollutants. This study examines the concurrent adsorption of phenol (PHE) and ciprofloxacin (CIP) onto three types of polyethylene terephthalate microplastics (PET MPs): pristine, acid-modified, and thermal-oxidatively aged. Using extended Langmuir (EL), extended Freundlich (EF) isotherms, and a new artificial neural network (ANN) model, equilibrium adsorption capacities were predicted. The EL isotherm fit for pristine and aged PET MPs, while EF fit for modified PET MPs. Monolayer adsorption capacities ranged from 342.10–3715.73 mg/g for PHE and 2518.23–14498.79 mg/g for CIP, exceeding single-component adsorption. The ANN model used one hidden layer with 3 neurons for pristine and aged PET MPs, and 2 hidden layers with five neurons for modified PET MPs, with a hyperbolic tangent activation function. Models showed excellent performance metrics, including R2 values of 0.989–0.999, RMSE of 0.001–0.413, and AAE of 0.009–0.327. Synergistic interactions were observed in the binary system, with PET MPs showing higher selectivity toward CIP. The study demonstrates the effectiveness of PET MPs for binary adsorption of PHE and CIP in aqueous solutions, highlighting their potential for multicomponent pollutant removal.

This study introduces a hybrid model that utilizes a model-based optimization method to generate training data and an artificial neural network (ANN)-based learning method to offer real-time exoskeleton support in lifting activities. For the model-based optimization method, the torque of the knee exoskeleton and the optimal lifting motion are predicted utilizing a two-dimensional (2D) human–exoskeleton model. The control points for exoskeleton motor current profiles and human joint angle profiles from cubic B-spline interpolation represent the design variables. Minimizing the square of the normalized human joint torque is considered as the cost function. Subsequently, the lifting optimization problem is tackled using a sequential quadratic programming (SQP) algorithm in sparse nonlinear optimizer (SNOPT). For the learning-based approach, the learning-based control model is trained using the general regression neural network (GRNN). The anthropometric parameters of the human subjects and lifting boundary postures are used as input parameters, while the control points for exoskeleton torque are treated as output parameters. Once trained, the learning-based control model can provide exoskeleton assistive torque in real time for lifting tasks. Two test subjects’ joint angles and ground reaction forces (GRFs) comparisons are presented between the experimental and simulation results. Furthermore, the utilization of exoskeletons significantly reduces activations of the four knee extensor and flexor muscles compared to lifting without the exoskeletons for both subjects. Overall, the learning-based control method can generate assistive torque profiles in real time and faster than the model-based optimal control approach.

Complex physical processes that are inherent to rainfall lead to the challenging task of its prediction. To contribute to the improvement of rainfall prediction, artificial neural network (ANN) models were developed using a multilayer perceptron (MLP) approach to predict monthly rainfall 2 months in advance for six geographically diverse weather stations across the Benin Republic. For this purpose, 12 lagged values of atmospheric data were used as predictors. The models were trained using data from 1959 to 2017 and tested for 4 years (2018–2021). The proposed method was compared to long short-term memory (LSTM) and climatology forecasts (CFs). The prediction performance was evaluated using five statistical measures: root mean square error, mean absolute error, mean absolute percentage error, coefficient of determination, and Nash–Sutcliffe efficiency (NSE) coefficient. Furthermore, Taylor diagrams, violin plots, box error, and Kruskal–Wallis test were used to assess the robustness of the model’s forecast. The results revealed that MLP gives better results than LSTM and CF. The NSE obtained with the MLP, LSTM, and CF models during the test period ranges from 0.373 to 0.885, 0.297 to 0.875, and 0.335 to 0.845, respectively, depending on the weather station. Rainfall predictability was more accurate, with 0.512 improvement in NSE using MLP at higher latitudes across the country, showing the effect of geographic regions on prediction model results. In summary, this research has revealed the potential of ANN techniques in predicting monthly rainfall 2 months ahead, supplying valuable insights for decision-makers in the Republic of Benin.

Real-time and efficient path planning is critical for all robotic systems. In particular, it is of greater importance for industrial robots since the overall planning and execution time directly impact the cycle time and automation economics in production lines. While the problem may not be complex in static environments, classical approaches are inefficient in high-dimensional environments in terms of planning time and optimality. Collision checking poses another challenge in obtaining a real-time solution for path planning in complex environments. To address these issues, we propose an end-to-end learning-based framework viz., Path Planning and Collision checking Network (PPCNet). The PPCNet generates the path by computing waypoints sequentially using two networks: the first network generates a waypoint, and the second one determines whether the waypoint is on a collision-free segment of the path. The end-to-end training process is based on imitation learning that uses data aggregation from the experience of an expert planner to train the two networks, simultaneously. We utilize two approaches for training a network that efficiently approximates the exact geometrical collision checking function. Finally, the PPCNet is evaluated in two different simulation environments and a practical implementation on a robotic arm for a bin-picking application. Compared to the state-of-the-art path-planning methods, our results show significant improvement in performance by greatly reducing the planning time with comparable success rates and path lengths.

In climate modeling, the stratospheric ozone layer is typically only considered in a highly simplified form due to computational constraints. For climate projections, it would be of advantage to include the mutual interactions between stratospheric ozone, temperature, and atmospheric dynamics to accurately represent radiative forcing. The overarching goal of our research is to replace the ozone layer in climate models with a machine-learned neural representation of the stratospheric ozone chemistry that allows for a particularly fast, but accurate and stable simulation. We created a benchmark data set from pairs of input and output variables that we stored from simulations of the ATLAS Chemistry and Transport Model. We analyzed several variants of multilayer perceptrons suitable for physical problems to learn a neural representation of a function that predicts 24-h ozone tendencies based on input variables. We performed a comprehensive hyperparameter optimization of the multilayer perceptron using Bayesian search and Hyperband early stopping. We validated our model by replacing the full chemistry module of ATLAS and comparing computation time, accuracy, and stability. We found that our model had a computation time that was a factor of 700 faster than the full chemistry module. The accuracy of our model compares favorably to the full chemistry module within a 2-year simulation run, also outperforms a previous polynomial approach for fast ozone chemistry, and reproduces seasonality well in both hemispheres. In conclusion, the neural representation of stratospheric ozone chemistry in simulation resulted in an ozone layer that showed a high accuracy, significant speed-up, and stability in a long-term simulation.

Hydroinformatics is a technology that combines information and communications technologies together with various disciplinary optimization and simulation models that focus on the management of water. This paper reviews the historical development of hydroinformatics and summarizes the current state of this technology. It describes the range of modeling tools and applications currently described in hydroinformatics literature. The paper concludes with some speculations about possible future developments in hydroinformatics.

In through-wall radar system, the wall parameters, including permittivity, and wall thickness are of crucial importance for locating targets precisely. Recently, to obtain a quick and accurate estimation of wall parameters, an approach based on machine learning was introduced. However, these approaches are less reliable as only simulations results are presented. One of the major concerns with machine learning-based approaches is the generation of training and testing data which require fabrication of wall with different permittivity, thickness, and conductivity. Creating walls with different permittivity, thickness, and conductivity can really be challenging and expensive. Therefore, an effort has been made in this paper to establish a cost-effective and robust machine learning-based wall parameter estimation process with the usage of transmission line method and artificial neural network. The implementation and efficacy of proposed approach have been demonstrated through simulation and experimental results. The proposed approach quickly and accurately predicted the wall relative permittivity and thickness of real building wall. The merit of proposed approach is that it is less complex and computational efficient as it can extract wall parameters from only one measurement and therefore can be used in conjunction with any commercial through-wall radar systems.

In this work, an artificial neural network model is established to understand the relationship among the tensile properties of as-printed Ti6Al4V parts, annealing parameters, and the tensile properties of annealed Ti6Al4V parts. The database was established by collecting published reports on the annealing treatment of selective laser melting (SLM) Ti6Al4V, from 2006 to 2020. Using the established model, it is possible to prescribe annealing parameters and predict properties after annealing for SLM Ti-6Al-4V parts with high confidence. The model shows high accuracy in the prediction of yield strength (YS) and ultimate tensile strength (UTS). It is found that the YS and UTS are sensitive to the annealing parameters, including temperature and holding time. The YS and UTS are also sensitive to initial YS and UTS of as-printed parts. The model suggests that an annealing process of the holding time of fewer than 4 h and the holding temperature lower than 850°C is desirable for as-printed Ti6Al4V parts to reach the YS required by the ASTM standard. By studying the collected data of microstructure and tensile properties of annealed Ti6Al4V, a new Hall-Petch relationship is proposed to correlate grain size and YS for annealed SLM Ti6Al4V parts in this work. The prediction of strain to failure shows lower accuracy compared with the predictions of YS and UTS due to the large scattering of the experimental data collected from the published reports.

Surface roughness (SR) is one of the major parameters used to govern the quality of the fused deposition modeling (FDM)-printed products, and the FDM process parameters can be easily regulated in order to obtain a good surface finish. The surface quality of the product produced by the FDM is generally affected by the staircase effect that needs to be managed. Also, the production time (PT) to fabricate the product and volume percentage error (VPE) should be minimized to make the FDM process more efficient. The aim of this paper is to accomplish these three objectives with the use of the parametric optimization technique integrating the artificial neural network (ANN) and the whale optimization algorithm (WOA). The FDM parameters which have been taken into consideration are layer thickness, nozzle temperature, printing speed, and raster width. Experimentation has been conducted on printed samples to examine the impact of the input parameters on SR, VPE, and PT according to Taguchi's L27 orthogonal array. The ANN model has been built up using the experimental data, which was further used as an objective function in the WOA with an aim to minimize output responses. The robustness of the proposed method has been validated on the optimal combinations of FDM process parameters.

In this paper, the design of frequency reconfigurable planar antenna by incorporation of metasurface superstrate (FRPA-MSS) is presented using an artificial neural network. The dual-layer radiating structure is created on a 1.524 mm thick Rogers RO4350B substrate board (εr = 3.48, tan δ = 0.0037). The candidate antenna is designed and analyzed using a high-frequency structure simulator (HFSS) tool. The transfer matrix method is employed for the successful retrieval of electromagnetic properties of the metamaterial. Frequency reconfiguration is achieved by placing the metasurface superstrate onto the rectangular patch antenna. A simplified ANN approach has been employed for the design of metasurface incorporated proposed antenna. Presented prototypes are characterized through experimental measurements. It is found from the practical observations that the proposed antenna effectively reconfigures the tuning range from 5.03 to 6.13 GHz. Moreover, the presented antenna operates efficiently with agreeable gain, good impedance matching, and stable pattern characteristics across the entire operational bandwidth. The experimental results obtained validate the simulated performance.

This study presents the main motivation to investigate the COVID-19 pandemic, a major threat to the whole world from the day when it first emerged in China city of Wuhan. Predictions on the number of cases of COVID-19 are crucial in order to prevent and control the outbreak. In this research study, an artificial neural network with rectifying linear unit-based technique is implemented to predict the number of deaths, recovered and confirmed cases of COVID-19 in Pakistan by using previous data of 137 days of COVID-19 cases from the day 25 February 2020 when the first two cases were confirmed, until 10 July 2020. The collected data were divided into training and test data which were used to test the efficiency of the proposed technique. Furthermore, future predictions have been made by the proposed technique for the next 7 days while training the model on whole available data.