Trained Neural Network Model Research Articles

The Mine Emergency Management Capability Based on EWM-CNN Comprehensive Evaluation

By establishing an evaluation system for emergency management capability for coal mine enterprises, the problems and shortfalls in coal mine emergency management can be identified. Moreover, this evaluation method holds promise in enhancing the ability to respond to emergencies in coal mines. The current research conducts both theoretical analysis and field research to determine and extract 26 factors that impact the emergency management capability of coal mining companies. The collected data is analyzed using SPSS version 24.0 and undergoes reliability and validity assessments. Subsequently, the entropy weight technique is utilized to determine the weight of individual evaluation system indices. As an effective evaluation tool, the trained convolutional neural network (CNN) model can accurately and effectively predict the satisfaction of the samples outside the training sample set, providing suggestions for improving coal mine emergency management. The trained model is employed to make empirical predictions and analyze the XY coal mine. The findings indicate that the overall emergency management capacity of the coal mine is II, which is at a good level. The evaluation results are consistent with the actual situation, and suggestions are provided for improving the emergency management ability of coal mines.

Non-Invasive Blood Pressure Sensing via Machine Learning.

In this paper, a machine learning (ML) approach to estimate blood pressure (BP) using photoplethysmography (PPG) is presented. The final aim of this paper was to develop ML methods for estimating blood pressure (BP) in a non-invasive way that is suitable in a telemedicine health-care monitoring context. The training of regression models useful for estimating systolic blood pressure (SBP) and diastolic blood pressure (DBP) was conducted using new extracted features from PPG signals processed using the Maximal Overlap Discrete Wavelet Transform (MODWT). As a matter of fact, the interest was on the use of the most significant features obtained by the Minimum Redundancy Maximum Relevance (MRMR) selection algorithm to train eXtreme Gradient Boost (XGBoost) and Neural Network (NN) models. This aim was satisfactorily achieved by also comparing it with works in the literature; in fact, it was found that XGBoost models are more accurate than NN models in both systolic and diastolic blood pressure measurements, obtaining a Root Mean Square Error (RMSE) for SBP and DBP, respectively, of 5.67 mmHg and 3.95 mmHg. For SBP measurement, this result is an improvement compared to that reported in the literature. Furthermore, the trained XGBoost regression model fulfills the requirements of the Association for the Advancement of Medical Instrumentation (AAMI) as well as grade A of the British Hypertension Society (BHS) standard.

A numerical control machining tool path step error prediction method based on BP neural network

Step error calculation of numerical control (NC) machining tool path is a premise for generating high-quality tool path and promoting its application. At present, iterative methods are generally used to calculate step error, and the computation time increases when accuracy improves. Neural networks can be calculated on GPUs and cloud platforms, which is conducive to reducing computation time and improving accuracy through continuous learning. This article innovatively introduces a BP neural network model to predict step error values. Firstly, the core parameters required for step error calculation are taken as the data samples to construct the neural network model, and map to the same scale through Z-score normalization to eliminate the adverse effects of singular parameters on the calculation results. Then, considering only a small number of parameters determine theoretical values of step error, the Dropout technique can drop hidden layer neurons with a certain probability, which is helpful to avoid overfitting and used in the neural network model design. In the neural network model training, this paper adds the Stochastic Gradient Descent with Momentum (SGDM) optimizer to the back propagation of network training in order to improves the network’ stability and accuracy. The proposed neural network predicts step error of samples from three surface models, the results show that the prediction error decreases as sample training increases. After trained by 15% of the surface samples, the neural network predicts the step errors of the remaining samples. Compared with theoretical values, more than 99% of the predicted values have an absolute error less than 1 μm. Moreover, the cost time is only one-third of the geometric method, which verifies the effectiveness and efficiency of our method.

Research on fast marking method for indicator diagram of pumping well based on K-means clustering

Indicator diagram is the key basis for fault diagnosis of pumping wells in oil exploitation. With the rapid development of machine learning, the fault diagnosis of indicator diagram based on deep learning has garnered increasing attention. This kind of methods train neural network models with marked samples, and then inputs images into the trained models and outputs their categories. At present, the preparation of indicator diagram sample set relies on experts' analysis of indicator diagram images one by one. However, it involves extensive manual work and manual marking is prone to errors, so the marked samples are often insufficient in quantity. In order to quickly mark a large number of indicator diagram samples, the oil well data was plotted into standardized indicator diagram, and then three feature extraction methods for indicator diagrams were proposed: feature extraction based on original vector, feature extraction based on three-dimensional pixel tensor, feature extraction based on convolutional neural network. These methods convert the indicator diagram into corresponding feature vectors, which are then clustered using the K-means clustering algorithm, enabling the corresponding indicator diagrams to be classified into different categories based on the clustering results. Using 20,000 randomly selected pieces of data from 100 pumping wells, this study clusters the sample set using the three proposed methods. The results indicated that the time consumption were 0.2, 8.3, and 0.7 h, with accuracy rates of 98%, 92%, and 95%, respectively. For indicator diagrams, the clustering method based on the original vector has outstanding performance in terms of efficiency and accuracy. This provides an automatic tool for the preparation of the pumping well fault diagnosis dataset, and its efficiency can be increased by tens of times compared with manual marking.

Survival prediction and prognostic factors in colorectal cancer after curative surgery: insights from cox regression and neural networks

Medical research frequently relies on Cox regression to analyze the survival distribution of cancer patients. Nonetheless, in specific scenarios, neural networks hold the potential to serve as a robust alternative. In this study, we aim to scrutinize the effectiveness of Cox regression and neural network models in assessing the survival outcomes of patients who have undergone treatment for colorectal cancer. We conducted a retrospective study on 284 colorectal cancer patients who underwent surgery at Imam Khomeini clinic in Hamadan between 2001 and 2017. The data was used to train both Cox regression and neural network models, and their predictive accuracy was compared using diagnostic measures such as sensitivity, specificity, positive predictive value, accuracy, negative predictive value, and area under the receiver operating characteristic curve. The analyses were performed using STATA 17 and R4.0.4 software. The study revealed that the best neural network model had a sensitivity of 74.5% (95% CI 61.0–85.0), specificity of 83.3% (65.3–94.4), positive predictive value of 89.1% (76.4–96.4), negative predictive value of 64.1% (47.2–78.8), AUC of 0.79 (0.70–0.88), and accuracy of 0.776 for death prediction. For recurrence, the best neural network model had a sensitivity of 88.1% (74.4–96.0%), specificity of 83.7% (69.3–93.2%), positive predictive value of 84.1% (69.9–93.4%), negative predictive value of 87.8% (73.8–95.9%), AUC of 0.86 (0.78–0.93), and accuracy of 0.859. The Cox model had comparable results, with a sensitivity of 73.6% (64.8–81.2) and 85.5% (78.3–91.0), specificity of 89.6% (83.8–93.8) and 98.0% (94.4–99.6), positive predictive value of 84.0% (75.6–90.4) and 97.4% (92.6–99.5), negative predictive value of 82.0% (75.6–90.4) and 88.8% (0.83–93.1), AUC of 0.82 (0.77–0.86) and 0.92 (0.89–0.95), and accuracy of 0.88 and 0.92 for death and recurrence prediction, respectively. In conclusion, the study found that both Cox regression and neural network models are effective in predicting early recurrence and death in patients with colorectal cancer after curative surgery. The neural network model showed slightly better sensitivity and negative predictive value for death, while the Cox model had better specificity and positive predictive value for recurrence. Overall, both models demonstrated high accuracy and AUC, indicating their usefulness in predicting these outcomes.

Forecast of pain degree of lumbar disc herniation based on back propagation neural network.

To further explore the pathogenic mechanism of lumbar disc herniation (LDH) pain, this study screens important imaging features that are significantly correlated with the pain score of LDH. The features with significant correlation imaging were included into a back propagation (BP) neural network model for training, including Pfirrmann classification, Michigan State University (MSU) regional localization (MSU protrusion size classification and MSU protrusion location classification), sagittal diameter index, sagittal diameter/transverse diameter index, transverse diameter index, and AN angle (angle between nerve root and protrusion). The BP neural network training model results showed that the specificity was 95 ± 2%, sensitivity was 91 ± 2%, and accuracy was 91 ± 2% of the model. The results show that the degree of intraspinal occupation of the intervertebral disc herniation and the degree of intervertebral disc degeneration are related to LDH pain. The innovation of this study is that the BP neural network model constructed in this study shows good performance in the accuracy experiment and receiver operating characteristic experiment, which completes the prediction task of lumbar Magnetic Resonance Imaging features for the pain degree of LDH for the first time, and provides a basis for subsequent clinical diagnosis.

Уточнення чисельних метеорологічних прогнозів за допомогою глибокого навчання

This paper briefly describes the history of numerical weather prediction development. The difficulties, which occur in the modelling of atmospheric processes, their nature and possible ways of their mitigation, are described. It also indicates alternative methods of improving the quality of meteorological forecasts. A brief history of deep learning and possible ways of its application to meteorological problems are given. Then, the paper describes the format used to store the 2m temperature forecasts of the COSMO numerical regional model. The proposed neural network architecture enables correcting the forecast errors of the numerical model. We conducted the experiments on the data of eight meteorological stations of the Kyiv region, so we obtained eight trained neural network models. The results showed that the proposed architecture enables obtaining better-quality forecasts in more than 50% of cases. Root-mean-square errors of the resulting forecasts decreased, and it is a widespread skill-score of improved-quality forecasts in meteorological science.

Design of oral English teaching model based on multi-modal perception of the Internet of Things and improved conventional neural networks.

Oral English instruction plays a pivotal role in educational endeavors. The emergence of online teaching in response to the epidemic has created an urgent demand for a methodology to evaluate and monitor oral English instruction. In the post-epidemic era, distance learning has become indispensable for educational pursuits. Given the distinct teaching modality and approach of oral English instruction, it is imperative to explore an intelligent scoring technique that can effectively oversee the content of English teaching. With this objective in mind, we have devised a scoring approach for oral English instruction based on multi-modal perception utilizing the Internet of Things (IoT). Initially, a trained convolutional neural network (CNN) model is employed to extract and quantify visual information and audio features from the IoT, reducing them to a fixed dimension. Subsequently, an external attention model is proposed to compute spoken English and image characteristics. Lastly, the content of English instruction is classified and graded based on the quantitative attributes of oral dialogue. Our findings illustrate that our scoring model for oral English instruction surpasses others, achieving the highest rankings and an accuracy of 88.8%, outperforming others by more than 2%.

Software implementation of the system for learning to write Chinese characters

Objective. The purpose of the work is devoted to the development and description of the mathematical model of the Chinese character recognition system, taking into account all the features of writing the Chinese language. The Chinese language learning app with character recognition module can help you replace a native speaker or home teacher for self-study. However, the developed software applications are based only on the creation of a neural network and cannot provide recognition, taking into account all the features of the language, which is so important when studying, therefore this topic is still relevant.Method. The neural network training model is based on the use of artificial neural networks using the backpropagation algorithm.Result. The article presents a software implementation of a system for teaching Chinese characters, taking into account the peculiarities of writing, the direction of each feature and its exact definition, taking into account the correct sequence and location in the character, as well as controlling the length of the features.Conclusion. Each of the writing features is an integral part of learning a language, since it can not only completely change the meaning of the written hieroglyph, but also help the learner to memorize the hieroglyph in a structured way, giving him a clear structure and algorithm of actions for writing the hieroglyph. When errors are detected when writing, the system will indicate to the user where exactly and in what area the error was made, what feature of the language he did not take into account, and attention should be paid to it.

DeMAC: Towards detecting model poisoning attacks in federated learning system

Federated learning (FL) is an efficient distributed machine learning paradigm for the collaborative training of neural network models by many clients with the assistance of a central server. Currently, the main challenging issue is that malicious clients can send poisoned model updates to the central server, making FL vulnerable to model poisoning attacks. In this paper, we propose a new system named DeMAC to improve the detection and defence against model poisoning attacks by malicious clients. The main idea behind the new system is based on an observation that, as malicious clients need to reduce the poisoning task learning loss, there will be obvious increases in the norm of gradients. We define a metric called GradScore to measure this norm of clients. It is shown through experiments that the GradScores of malicious and benign clients are distinguishable in all training stages. Therefore, DeMAC can detect malicious clients by measuring the GradScore of clients. Furthermore, a historical record of the contributed global model updates is utilized to enhance the DeMAC system, which can spontaneously detect malicious behaviours without requiring manual settings. Experiment results over two benchmark datasets show that DeMAC can reduce the attack success rate under various attack strategies. In addition, DeMAC can eliminate model poisoning attacks under heterogeneous environments.

A Model for Yield Estimation Based on Sea Buckthorn Images

Sea buckthorn is an extremely drought-tolerant, resilient and sustainable crop that can be grown in areas with harsh climates and scarce resources to provide a source of nutrition and income for the local population. The use of image-based yield estimation methods allows for better management of sea buckthorn cultivation to improve its productivity and sustainability, while the error in fruit yield information due to occlusion can be well reduced by combining and analysing the image features extracted using binocular cameras. In this paper, mature wild sea buckthorn in the mountainous areas north of Hohhot City, Inner Mongolia Autonomous Region, were used as the study target. Firstly, complete images of sea buckthorn branches were collected by binocular cameras and features were extracted. The extracted features include the colour index of sea buckthorn fruits, the number of fruits and a total of four texture parameters, ASM, CON, COR and HOM. The features with significant correlation to sea buckthorn fruit weight were selected by correlation calculation of the feature parameters, the obtained correlation features were introduced into the BP neural network model for training and then the sea buckthorn estimation model was obtained. The results showed that the best yield estimation model was achieved by combining the COR index with the colour index and the number of sea buckthorn fruits, with a coefficient of determination R2 = 0.99267 and a root mean square error RMSE = 0.5214.

Toward a general neural network force field for protein simulations: Refining the intramolecular interaction in protein.

Molecular dynamics (MD) is an extremely powerful, highly effective, and widely used approach to understanding the nature of chemical processes in atomic details for proteins. The accuracy of results from MD simulations is highly dependent on force fields. Currently, molecular mechanical (MM) force fields are mainly utilized in MD simulations because of their low computational cost. Quantum mechanical (QM) calculation has high accuracy, but it is exceedingly time consuming for protein simulations. Machine learning (ML) provides the capability for generating accurate potential at the QM level without increasing much computational effort for specific systems that can be studied at the QM level. However, the construction of general machine learned force fields, needed for broad applications and large and complex systems, is still challenging. Here, general and transferable neural network (NN) force fields based on CHARMM force fields, named CHARMM-NN, are constructed for proteins by training NN models on 27 fragments partitioned from the residue-based systematic molecular fragmentation (rSMF) method. The NN for each fragment is based on atom types and uses new input features that are similar to MM inputs, including bonds, angles, dihedrals, and non-bonded terms, which enhance the compatibility of CHARMM-NN to MM MD and enable the implementation of CHARMM-NN force fields in different MD programs. While the main part of the energy of the protein is based on rSMF and NN, the nonbonded interactions between the fragments and with water are taken from the CHARMM force field through mechanical embedding. The validations of the method for dipeptides on geometric data, relative potential energies, and structural reorganization energies demonstrate that the CHARMM-NN local minima on the potential energy surface are very accurate approximations to QM, showing the success of CHARMM-NN for bonded interactions. However, the MD simulations on peptides and proteins indicate that more accurate methods to represent protein-water interactions in fragments and non-bonded interactions between fragments should be considered in the future improvement of CHARMM-NN, which can increase the accuracy of approximation beyond the current mechanical embedding QM/MM level.

BRoCoM: A Bayesian Framework for Robust Computing on Memristor Crossbar

Memristor crossbar arrays are considered to be a promising platform for neuromorphic computing. To deploy a trained neural network model on memristor crossbars, memristors need to be programmed to the corresponding weight values. In fact, due to device based process variation and noise, deviations of the stored weights from the trained weights are inevitable, thereby causing the degradation of the actual inference performance. This paper proposes a unified Bayesian inference based framework, BRoCoM, which connects device nonidealities and algorithmic training together for robust computing on memristor crossbars. BRoCoM is able to incorporate different levels of nonidealities into prior weight distribution, and transform robustness optimization to Bayesian neural network training, the weights of neural networks are optimized to accommodate uncertainties and minimize inference degradation. Experimental results confirm the capability of the proposed BRoCoM to achieve stable inference performance while tolerating the non-ideal effects of process variation and noise.

Research on Prediction of User's Entrepreneurial Learning Behavior based on Neural Network

This study aims to explore the predictive capability of neural networks in user entrepreneurial learning behavior. Traditional methods have limitations in predicting user entrepreneurial learning behavior. Therefore, neural networks are adopted as predictive models, analyzing multidimensional data such as user personal information, learning history, and learning behavior. The research methodology includes data collection and preprocessing, feature engineering, design and training of neural network models, as well as evaluation and interpretation of prediction results. Through experimentation and result analysis, the neural network models demonstrate good accuracy and predictive ability in forecasting user entrepreneurial learning behavior. The research results further explain the key factors influencing user entrepreneurial learning behavior and provide insights for improving entrepreneurial education and learning platforms. This study offers new methods and theoretical foundations for predicting user entrepreneurial learning behavior, significantly contributing to enhancing learning outcomes and entrepreneurial success rates for entrepreneurs.

Structural Assessment Based on Vibration Measurement Test Combined with an Artificial Neural Network for the Steel Truss Bridge

Damage assessment is one of the most crucial issues for bridge engineers during the operational and maintenance phase, especially for existing steel bridges. Among several methodologies, the vibration measurement test is a typical approach, in which the natural frequency variation of the structure is monitored to detect the existence of damage. However, locating and quantifying the damage is still a big challenge for this method, due to the required human resources and logistics involved. In this regard, an artificial intelligence (AI)-based approach seems to be a potential way of overcoming such obstacles. This study deployed a comprehensive campaign to determine all the dynamic parameters of a predamaged steel truss bridge structure. Based on the results for mode shape, natural frequency, and damping ratio, a finite element model (FEM) was created and updated. The artificial intelligence network’s input data from the damage cases were then analysed and evaluated. The trained artificial neural network model was curated and evaluated to confirm the approach’s feasibility. During the actual operational stage of the steel truss bridge, this damage assessment system showed good performance, in terms of monitoring the structural behaviour of the bridge under some unexpected accidents.

A trained neural network model accurately predicts Achilles tendon stress during walking and running based on shear wave propagation

Shear wave tensiometry is a noninvasive technique for measuring tendon loading during activity based on the speed of a shear wave traveling along the tendon. Shear wave speed has been shown to modulate with axial stress, but calibration is required to obtain absolute measures of tendon loading. However, the current technique only makes use of wave speed, whereas other characteristics of the wave (e.g., amplitude, frequency content) may also vary with tendon loading. It is possible that these data could be used in addition to wave speed to circumvent the need for calibration. Given the potential complex relationships to tendon loading, and the lack of an analytical model to guide the use of these data, it is sensible to use a machine learning approach. Here, we used an ensemble neural network approach to predict inverse dynamics estimates of Achilles tendon stress from shear wave tensiometry data collected in a prior study. Neural network-predicted stresses were highly correlated with stance phase inverse dynamics estimates for walking (R2 = 0.89 ± 0.06) and running (R2 = 0.87 ± 0.11) data reserved for neural network model testing and not included in model training. Additionally, error between neural network-predicted and inverse dynamics-estimated stress was reasonable (walking: RMSD = 11 ± 2% of peak load; running: 25 ± 14%). Results of this pilot analysis suggest that a machine learning approach could reduce the reliance of shear wave tensiometry on calibration and expand its usability in many settings.

Presenting the Self-Healing Potential Model by Induction Heating Method in Warm Asphalt Mixtures by Artificial Neural Network

The increasing number of vehicles in the world and the increase in the dimensions of light and heavy vehicles with different combinations, including simple and compound axles, have caused various types of damage in the asphalt, which affects the performance of the roads and requires a suitable pavement. For this reason, engineers try to improve the performance of asphalt mixture in different ways and increase their useful life. Cracking is one of the main failure modes of asphalt pavements. Fatigue cracking is the most common form of asphalt failure and is one of the important factors in reducing the life of road pavements. The purpose of this research is to investigate the index of self-healing potential (HI) of asphalt mixtures containing activated carbon and zycotherm by induction heating method and modelling by artificial neural network. First, by making warm mixture asphalt samples containing activated carbon, self-healing by induction heating method was provided in the pavement, and then the effect of traffic loading and environmental conditions simultaneously in terms of changes in loading parameters, taking into account the conditions of the service life. The results show that the use of activated carbon increases the values of shear modulus. It also improves the properties of modified bitumen at all temperatures. The resistance of the asphalt mixture against breaking and cracking, as well as the maximum force tolerance at the temperature of -16°C is higher than at the temperature of 25°C, which is one of the reasons for the reduction of flexibility of bitumen at low temperature. Also, for modelling, perceptron multi-layer neural network (MLP) with 9 input variables, three hidden layers and one output layer has been used. The results showed that the explanation coefficient of the neural network model for training, validation and evaluation data was 0.99717 for training, 0.62053 for validation and 0.9301 for testing. These values correctly show the high fitting ability of the proposed neural network model. Also, the lowest value obtained for the mean squared error is 4.9223x10^(-5) and in epoch 14.

A Weight-Generating Approach of a Deep Neural Network for the Parameter Identification of Dynamic Systems

The general learning process of deep learning is extremely time-consuming. Unlike the traditional learning process, a weight-generating approach to quickly generate the weight vectors of a deep neural network model is proposed, which can be used for parameter identification of a dynamic system. Based on the analysis of three trained deep neural network models, which are used to identify the parameters of three different dynamic systems, the statistical relationships between the weight vectors of each hidden layer and its inputs are revealed. Then, the statistical patterns of the weight vectors are imitated by exploiting the statistical patterns of the inputs and these relationships. Then, a weight-generating approach is designed to quickly generate the weight vectors of a deep neural network model. The effectiveness of the weight-generating approach is tested on the tasks of parameter identification for the three dynamic systems. The numerical results are provided to demonstrate the validity and high efficiency of the proposed weight-generating approach.

Training Neural Network Models Using Molecular Dynamics Simulation Results to Efficiently Predict Cyclic Hexapeptide Structural Ensembles.

Cyclic peptides have emerged as a promising class of therapeutics. However, their de novo design remains challenging, and many cyclic peptide drugs are simply natural products or their derivatives. Most cyclic peptides, including the current cyclic peptide drugs, adopt multiple conformations in water. The ability to characterize cyclic peptide structural ensembles would greatly aid their rational design. In a previous pioneering study, our group demonstrated that using molecular dynamics results to train machine learning models can efficiently predict structural ensembles of cyclic pentapeptides. Using this method, which was termed StrEAMM (Structural Ensembles Achieved by Molecular Dynamics and Machine Learning), linear regression models were able to predict the structural ensembles for an independent test set with R2 = 0.94 between the predicted populations for specific structures and the observed populations in molecular dynamics simulations for cyclic pentapeptides. An underlying assumption in these StrEAMM models is that cyclic peptide structural preferences are predominantly influenced by neighboring interactions, namely, interactions between (1,2) and (1,3) residues. Here we demonstrate that for larger cyclic peptides such as cyclic hexapeptides, linear regression models including only (1,2) and (1,3) interactions fail to produce satisfactory predictions (R2 = 0.47); further inclusion of (1,4) interactions leads to moderate improvements (R2 = 0.75). We show that when using convolutional neural networks and graph neural networks to incorporate complex nonlinear interaction patterns, we can achieve R2 = 0.97 and R2 = 0.91 for cyclic pentapeptides and hexapeptides, respectively.

Removing the cardiac field artifact from the EEG using neural network regression.

When EEG recordings are used to reveal interactions between central-nervous and cardiovascular processes, the cardiac field artifact (CFA) poses a major challenge. Because the electric field generated by cardiac activity is also captured by scalp electrodes, the CFA arises as a heavy contaminant whenever EEG data are analyzed time-locked to cardio-electric events. A typical example is measuring stimulus-evoked potentials elicited at different phases of the cardiac cycle. Here, we present a nonlinear regression method deploying neural networks that allows to remove the CFA from the EEG signal in such scenarios. We train neural network models to predict R-peak centered EEG episodes based on the ECG and additional CFA-related information. In a second step, these trained models are used to predict and consequently remove the CFA in EEG episodes containing visual stimulation occurring time-locked to the ECG. We show that removing these predictions from the signal effectively removes the CFA without affecting the intertrial phase coherence of stimulus-evoked activity. In addition, we provide the results of an extensive grid search suggesting a set of appropriate model hyperparameters. The proposed method offers a replicable way of removing the CFA on the single-trial level, without affecting stimulus-related variance occurring time-locked to cardiac events. Disentangling the cardiac field artifact (CFA) from the EEG signal is a major challenge when investigating the neurocognitive impact of cardioafferent traffic by means of the EEG. When stimuli are presented time-locked to the cardiac cycle, both sources of variance are systematically confounded. Here, we propose a regression-based approach deploying neural network models to remove the CFA from the EEG. This approach effectively removes the CFA on a single-trial level and is purely data-driven, providing replicable results.