Generalization Capability Research Articles

Deep learning in motor imagery EEG signal decoding: A Systematic Review

Thanks to the fast evolution of electroencephalography (EEG)-based brain–computer interfaces (BCIs) and computing technologies, as well as the availability of large EEG datasets, decoding motor imagery (MI) EEG signals is rapidly shifting from traditional machine learning (ML) to deep learning (DL) approaches. Furthermore, real-world MI-EEG BCI applications are progressively requiring higher generalization capabilities, which can be achieved by leveraging publicly available MI-EEG datasets and high-performance decoding models. Within this context, this paper provides a systematic review of DL approaches for MI-EEG decoding, focusing on studies that work on publicly available EEG-MI datasets. This review paper firstly provides a clear overview of these datasets that can be used for DL model training and testing. Afterwards, considering each dataset, related DL studies are discussed with respect to the four decoding paradigms identified in the literature, i.e., subject-dependent, subject-independent, transfer learning, and global decoding paradigms. Having analyzed the reviewed studies, the current trends and strategies, popular architectures, baseline models that are used for comprehensive analysis, and techniques to ensure reproducibility of the results in DL-based MI-EEG decoding are also identified and discussed. The selection and screening of the studies included in this review follow the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, leading to a comprehensive analysis of 394 papers published between January 1, 2017, and January 23, 2023.

Explainable Encoder-Prediction-Reconstruction Framework for the Prediction of Metasurface Absorption Spectra.

The correlation between metasurface structures and their corresponding absorption spectra is inherently complex due to intricate physical interactions. Additionally, the reliance on Maxwell's equations for simulating these relationships leads to extensive computational demands, significantly hindering rapid development in this area. Numerous researchers have employed artificial intelligence (AI) models to predict absorption spectra. However, these models often act as black boxes. Despite training high-performance models, it remains challenging to verify if they are fitting to rational patterns or merely guessing outcomes. To address these challenges, we introduce the Explainable Encoder-Prediction-Reconstruction (EEPR) framework, which separates the prediction process into feature extraction and spectra generation, facilitating a deeper understanding of the physical relationships between metasurface structures and spectra and unveiling the model's operations at the feature level. Our model achieves a 66.23% reduction in average Mean Square Error (MSE), with an MSE of 2.843 × 10-4 compared to the average MSE of 8.421×10-4 for mainstream networks. Additionally, our model operates approximately 500,000 times faster than traditional simulations based on Maxwell's equations, with a time of 3×10-3 seconds per sample, and demonstrates excellent generalization capabilities. By utilizing the EEPR framework, we achieve feature-level explainability and offer insights into the physical properties and their impact on metasurface structures, going beyond the pixel-level explanations provided by existing research. Additionally, we demonstrate the capability to adjust absorption by changing the metasurface at the feature level. These insights potentially empower designers to refine structures and enhance their trust in AI applications.

HybridCBAMNet: Enhancing time series binary classification with convolutional recurrent networks and attention mechanisms

The rapid advancement of Internet of Things technology and the increasing availability of big data have resulted in an exponential growth of time series data, highlighting a pressing need for effective classification methods. This study introduces HybridCBAMNet, a novel convolutional recurrent neural network model enhanced with recurrent networks and attention mechanisms for binary time series classification. The architecture integrates Conv1D-based feature extraction modules to extract relevant features, alongside attention enhancement modules and convolutional block attention modules. Additionally, bidirectional recurrent units are utilized to capture temporal dependencies and contextual information in both forward and backward directions. The model achieves top F1-scores in seven out of thirty-six binary classification tasks, significantly surpassing the performance of fourteen existing state-of-the-art models from the UCR archive. These results demonstrate that HybridCBAMNet not only enhances classification accuracy but also improves the model generalization capabilities, contributing valuable insights to the field of time series analysis.

A Hybrid System Based on Bayesian Networks and Deep Learning for Explainable Mental Health Diagnosis

Mental illnesses are becoming one of the most common health concerns among the population. Despite the proven efficacy of psychological treatments, mental illnesses are largely underdiagnosed, particularly in developing countries. A key factor contributing to this is the scarcity of mental health providers capable of diagnosing. In this work, we propose a novel method that combines the general capabilities and accuracy of Large Language models with the explainability of Bayesian Networks. Our system analyzes descriptions of symptoms provided by users and written in natural language and, based on these descriptions, asks questions to confirm or refine the initial diagnosis made by the deep learning model. We trained our model on a large-scale dataset collected from various internet sources, comprising over 2.3 million data points. The initial prediction from the Large Language model is refined through symptom confirmation questions derived from a probabilistic graphical model constructed by experts based on the DSM-5 diagnostic manual. We present results from symptom descriptions sourced from the internet and clinical vignettes extracted from behavioral science exams, demonstrating the effectiveness of our hybrid model in classifying mental health disorders. Our model achieves high accuracy in classifying a wide range of mental health disorders, providing transparent and explainable predictions.

Research on Fault Diagnosis of Rolling Bearing Based on Gramian Angular Field and Lightweight Model.

Due to the limitations of deep learning models in processing one-dimensional signal feature extraction, and high model complexity leading to low training accuracy and large consumption of computing resources, this paper innovatively proposes a rolling bearing fault diagnosis method based on Gramian Angular Field (GAF) and enhanced lightweight residual network. Firstly, the one-dimensional signal is transformed into a two-dimensional GAF image, fully preserving the signal's temporal dependency. Secondly, to address the parameter redundancy and high computational complexity of the ResNet-18 model, its residual blocks are improved. The second convolutional layer in the downsampling residual blocks is removed, traditional convolutional layers are replaced with depthwise separable convolutions, and the lightweight Efficient Channel Attention (ECA) module is embedded after each residual block. This further enhances the model's ability to capture key features while maintaining low computational cost, resulting in a lightweight model referred to as E-ResNet13. Finally, the generated GAF feature maps are fed into the E-ResNet13 model for training, and through a global average pooling layer, they are mapped to a fully connected layer for classifying the faults of rolling bearings. Verifying the superiority of the proposed GAF-E-ResNet13 model, experimental results show that the GAF image encoding method achieves higher fault recognition accuracy compared to other encoding methods. Compared with other intelligent diagnosis methods, the E-ResNet13 model demonstrates strong diagnostic performance and generalization capability under both a single condition and complex varying conditions, fully proving the innovation and practicality of this method.

LG-AKD: Application of a lightweight GCN model based on adversarial knowledge distillation to skeleton action recognition

Human action recognition, a pivotal topic in computer vision, is a highly complex and challenging task. It requires the analysis of not only spatial dependencies of targets but also temporal changes in these targets. In recent decades, the advancement of deep learning has led to the development of numerous action recognition methods based on deep neural networks. Given that the skeleton points of the human body can be treated as a graph structure, graph neural networks (GNNs) have emerged as an effective tool for modeling such data, garnering significant interest from researchers. This paper aims to address the issue of low test speed caused by over-complicated deep graph convolutional models. To achieve this, we compress the network structure using knowledge distillation from a teacher-student architecture, leading to a compact and lightweight student GNN. To enhance the model’s robustness and generalization capabilities, we introduce a data augmentation mechanism that generates diverse action sequences while maintaining consistent behavior labels, thereby providing a more comprehensive learning basis for the model. The proposed model integrates three distinct knowledge learning paths: teacher networks, original datasets, and derived data. The fusion of knowledge distillation and data augmentation enables lightweight student networks to outperform their teacher networks in terms of both performance and efficiency. Experimental results demonstrate the efficacy of our approach in the context of skeleton-based human action recognition, highlighting its potential to simplify state-of-the-art models while enhancing their performance.

A novel hybrid BWO-BiLSTM-ATT framework for accurate offshore wind power prediction

Accurate wind power forecasting is pivotal in facilitating the efficient integration of renewable energy sources into existing power grids. This study proposes a novel hybrid framework, termed BWO-BiLSTM-ATT, which synergistically combines the strengths of a bidirectional long short-term memory (BiLSTM) network, a self-attention mechanism, and the Beluga Whale Optimization (BWO) algorithm. The BiLSTM architecture, with its unique ability to capture bidirectional temporal dependencies, effectively models the intricate dynamics present in wind power time series data. Integrating the self-attention mechanism further enhances the model's performance by discerning and emphasizing the most salient features within the input data. Furthermore, employing the BWO algorithm optimizes the model's hyperparameters, ensuring optimal configuration and enhancing its predictive accuracy and generalization capabilities. The proposed BWO-BiLSTM-ATT framework was rigorously evaluated using real-world data from an offshore wind farm in Yangjiang City, China. Comparative analyses were conducted against several baseline algorithms, including persistence, ARIMA, SVR, and vanilla LSTM models. The results demonstrate the superior performance of the BWO-BiLSTM-ATT framework, achieving higher R-squared and explained variance scores, coupled with lower error metrics such as MSE, RMSE, MedAE, and MAE. Statistical significance tests further corroborated the substantial performance improvements offered by the proposed framework. The combination of advanced deep learning techniques and metaheuristic optimization algorithms embedded in the BWO-BiLSTM-ATT framework provides a robust and accurate solution for wind power forecasting, enabling efficient management and seamless integration of renewable energy resources into existing power grids.

MFR-Net: A multi-feature fusion phase unwrapping method for different speckle noises

Phase unwrapping is a crucial step in laser interferometry for obtaining accurate physical measurement of object. To reduce the impact of speckle noise on wrapped phase during actual measurement and improve the subsequent measurement accuracy, a multi-feature fusion phase unwrapping method for different speckle noises named MFR-Net is proposed in this paper. The network is composed of a front-end multi-module filter processing layer and a back-end network with dilated convolution and coordinate attention mechanism. By reducing random phase differences introduced by different levels of noise, the network enhances its capability to extract spatial features such as gradient information between pixels under speckle noise, so that it successfully unwraps the wrapped phase with different speckle noises and accurately recovers the real phase information. Taking the wrapped phases with multiplicative speckle noise and additive random noise as dataset, the results of ablation and comparison experiments show that the MFR-Net has superior unwrapped results. Under three different levels of speckle noise, the average values of MSE, SSIM, PSNR and AU for MFR-Net are at least improved by 84.80 %, 10.99 %, 29.00 % and 7.72 %, respectively, compared to PDVQG, TIE, DLPU and VURNet algorithms. When the standard deviation of speckle noise varies continuously in the range [1.0, 2.0], the average values of four indexes reaches 0.12 rad, 0.91, 31.80 dB and 99.96 %, respectively, indicating the stronger robustness of MFR-Net. In addition, the phase step unwrapping is performed by MFR-Net. Compared to DLPU and VURNet, MFR-Net method reduced MSE by 80 % and 87.35 %, respectively, demonstrating the outstanding generalization capability. The proposed MFR-Net can realize the correct phase unwrapping under different speckle noises. It may be applied in laser interferometry applications such as digital holography and interferometric synthetic aperture radar.

Tree-based machine learning for predicting Neochloris oleoabundans biomass growth and biological nutrient removal from tertiary municipal wastewater

Recently, computational models have been increasingly recognized as valuable tools for addressing key challenges in the operational performance of biological wastewater treatment facilities. In this study, tree-based machine learning approaches, such as decision tree regressor (DTR) and extra tree regressor (ETR), were developed to predict microalgae (Neochloris oleoabundans) biomass growth, culture pH, and nutrient removal efficacy (total nitrogen, TN and total phosphorus, TP) for the first time. The experimental data was obtained through a central composite design (CCD) matrix, and Bayesian optimization was applied to fine-tune the models’ hyperparameters. Model performance was evaluated using indicators such as the coefficient of determination (R²), mean absolute error (MAE), and mean-squared error (MSE). The results showed comparable performance between the DTR and ETR models. For TN removal during testing, the R² values for DTR and ETR were 0.9262 and 0.9789, respectively, with DTR (MSE: 0.00895, MAE: 0.0615) and ETR (MSE: 0.00255, MAE: 0.0352) demonstrating reliable predictions. Overall, the ETR model outperformed DTR in predicting responses. The models' generalization capabilities were also assessed by introducing variations in environmental factors.

Bilateral Proxy Federated Domain Generalization for Privacy-preserving Medical Image Diagnosis.

Contemporary domain generalization methods have demonstrated effectiveness in aiding the generalized diagnosis of medical images with multi-source data by joint optimization. However, the centralized training paradigm employed by these approaches becomes infeasible when data are non-shared across domains due to the high privacy of medical data. Despite attempts by existing federated domain generalization methods to address this issue, the simultaneous attainment of strict privacy protection and a satisfactory level of generalization ability on out-of-distribution data remains a persistent challenge. In this paper, to tackle this challenging problem, we propose a novel approach called the Bilateral Proxy Framework (BPF). The BPF leverages the client-side proxies to facilitate the strict privacy-preserving communications with the server and ensure smoother and more stable convergences of local models through mutual distillation. Meanwhile, the server-side proxy adopts a distance-based strategy and a parameter moving average scheme, which enhances the stability and robustness of the global model, particularly by averting abrupt parameter changes that could result in fluctuations or overfitting. Through these advancements, our framework strives to enhance the generalization capability of the global model, enabling more accurate and reliable medical image diagnosis in federated settings. The effectiveness of our method is demonstrated with superior performance over state-of-the-arts on both simulated and real-world distribution medical image diagnosis tasks.

A novel multi-modal fusion method based on uncertainty-guided meta-learning

Multi-modal data fusion for effective feature representation in machine learning is challenging due to intrinsic biases present within and across different modalities. Existing multi-modal data fusion methods often face difficulties in learning generic features due to diverse noise patterns and variations in feature dynamics across different modalities. In this paper, we present a novel method called Uncertainty-guided Meta-Learning Multi-modal Fusion and Classification (UMLMC) to address these challenges. UMLMC dynamically transforms multi-modal feature spaces at both the pre- and post-fusion levels by incorporating uncertainty estimates from an auxiliary network. Our model is optimized using a meta-learning algorithm to enhance its generalization capabilities. Extensive experiments on multi-modal data from diverse domains, along with comparisons to state-of-the-art methods, demonstrate the effectiveness of UMLMC in improving classification performance. These results confirm that UMLMC, with its innovative uncertainty estimation and meta-learning framework, effectively learns informative intra- and inter-modal features, leading to superior classification outcomes.

Subject-Independent Wearable P300 Brain-Computer Interface Based on Convolutional Neural Network and Metric Learning.

The calibration procedure for a wearable P300 brain-computer interface (BCI) greatly impact the user experience of the system. Each user needs to spend additional time establishing a decoder adapted to their own brainwaves. Therefore, achieving subject independent is an urgent issue for wearable P300 BCI needs to be addressed. A dataset of electroencephalogram (EEG) signals was constructed from 100 individuals by conducting a P300 speller task with a wearable EEG amplifier. A framework is proposed that initially improves cross-subject consistency of EEG features through a common feature extractor. Subsequently, a simple and compact convolutional neural network (CNN) architecture is employed to learn an embedding sub-space, where the mapped EEG features are maximally separated, while pursuing the minimum distance within the same class and the maximum distance between different classes. Finally, the model's generalization capability was further optimized through fine-tuning. The proposed method significantly boosts the average accuracy of wearable P300 BCI to 73.23 ±7.62% without calibration and 78.75±6.37% with fine-tuning. The results demonstrate the feasibility and excellent performance of our dataset and framework. A calibration-free wearable P300 BCI system is feasible, suggesting significant potential for practical applications of the wearable P300 BCI system.

HBANet: A hybrid boundary-aware attention network for infrared and visible image fusion

Infrared and visible image fusion is an extensively investigated problem in infrared image processing, aiming to extract useful information from source images. However, the automatic fusion of these images presents a significant challenge due to the large domain difference and ambiguous boundaries. In this article, we propose a novel image fusion approach based on hybrid boundary-aware attention, termed HBANet, which models global dependencies across the image and leverages boundary-wise prior knowledge to supplement local details. Specifically, we design a novel mixed boundary-aware attention module that is capable of leveraging spatial information to the fullest extent and integrating long dependencies across different domains. To preserve the integrity of texture and structural information, we introduced a sophisticated loss function that comprises structure, intensity, and variation losses. Our method has been demonstrated to outperform state-of-the-art methods in terms of both visual and quantitative metrics, in our experiments on public datasets. Furthermore, our approach also exhibits great generalization capability, achieving satisfactory results in CT and MRI image fusion tasks.

Estimation of daily XCO2 at 1 km resolution in China using a spatiotemporal ResNet model

Carbon dioxide (CO2) serves as a crucial greenhouse gas that traps heat and regulates the Earth's temperature. High spatiotemporal resolution CO2 estimation can provide valuable information to understand the characteristics of fine-scale climate change trends and to formulate more effective emission reduction strategies. This study presents a spatiotemporal ResNet model (ST-ResNet) specifically developed to estimate the highest resolution (1 km × 1 km) daily column-averaged dry-air mole fraction of CO2 (XCO2) in China from 2015 to 2020. The ST-ResNet model excels in estimating XCO2 by comprehensively considering the complex relationships between XCO2 and its various influencing factors, while efficiently capturing both temporal and spatial correlations, thereby demonstrating remarkable generalization capability. The results show that the ST-ResNet generates a highly accurate XCO2 dataset, outperforming the traditional ResNet. Ground-based validation results further confirm the high accuracy and spatiotemporal resolution of our estimated data product. Using this dataset, the spatial and temporal characteristics of XCO2 across the entire China and several urban agglomerations have been analyzed. The high spatiotemporal resolution estimated XCO2 dataset for China is made publicly available at [https://doi.org/10.6084/m9.figshare.25272868], offering substantial potential for fine-scale carbon research.

Reinforcement learning as a robotics-inspired framework for insect navigation: from spatial representations to neural implementation.

Bees are among the master navigators of the insect world. Despite impressive advances in robot navigation research, the performance of these insects is still unrivaled by any artificial system in terms of training efficiency and generalization capabilities, particularly considering the limited computational capacity. On the other hand, computational principles underlying these extraordinary feats are still only partially understood. The theoretical framework of reinforcement learning (RL) provides an ideal focal point to bring the two fields together for mutual benefit. In particular, we analyze and compare representations of space in robot and insect navigation models through the lens of RL, as the efficiency of insect navigation is likely rooted in an efficient and robust internal representation, linking retinotopic (egocentric) visual input with the geometry of the environment. While RL has long been at the core of robot navigation research, current computational theories of insect navigation are not commonly formulated within this framework, but largely as an associative learning process implemented in the insect brain, especially in the mushroom body (MB). Here we propose specific hypothetical components of the MB circuit that would enable the implementation of a certain class of relatively simple RL algorithms, capable of integrating distinct components of a navigation task, reminiscent of hierarchical RL models used in robot navigation. We discuss how current models of insect and robot navigation are exploring representations beyond classical, complete map-like representations, with spatial information being embedded in the respective latent representations to varying degrees.

Crowd Counting Using Meta-Test-Time Adaptation.

Machine learning algorithms are commonly used for quickly and efficiently counting people from a crowd. Test-time adaptation methods for crowd counting adjust model parameters and employ additional data augmentation to better adapt the model to the specific conditions encountered during testing. The majority of current studies concentrate on unsupervised domain adaptation. These approaches commonly perform hundreds of epochs of training iterations, requiring a sizable number of unannotated data of every new target domain apart from annotated data of the source domain. Unlike these methods, we propose a meta-test-time adaptive crowd counting approach called CrowdTTA, which integrates the concept of test-time adaptation into the meta-learning framework and makes it easier for the counting model to adapt to the unknown test distributions. To facilitate the reliable supervision signal at the pixel level, we introduce uncertainty by inserting the dropout layer into the counting model. The uncertainty is then used to generate valuable pseudo labels, serving as effective supervisory signals for adapting the model. In the context of meta-learning, one image can be regarded as one task for crowd counting. In each iteration, our approach is a dual-level optimization process. In the inner update, we employ a self-supervised consistency loss function to optimize the model so as to simulate the parameters update process that occurs during the test phase. In the outer update, we authentically update the parameters based on the image with ground truth, improving the model's performance and making the pseudo labels more accurate in the next iteration. At test time, the input image is used for adapting the model before testing the image. In comparison to various supervised learning and domain adaptation methods, our results via extensive experiments on diverse datasets showcase the general adaptive capability of our approach across datasets with varying crowd densities and scales.

Interval Constrained Multi-Objective Optimization Scheduling Method for Island-Integrated Energy Systems Based on Meta-Learning and Enhanced Proximal Policy Optimization

Multiple uncertainties from source–load and energy conversion significantly impact the real-time dispatch of an island integrated energy system (IIES). This paper addresses the day-ahead scheduling problems of IIES under these conditions, aiming to minimize daily economic costs and maximize the output of renewable energies. We introduce an innovative algorithm for Interval Constrained Multi-objective Optimization Problems (ICMOPs), which incorporates meta-learning and an improved Proximal Policy Optimization with Clipped Objective (PPO-CLIP) approach. This algorithm fills a notable gap in the application of DRL to complex ICMOPs within the field. Initially, the multi-objective problem is decomposed into several single-objective problems using a uniform weight decomposition method. A meta-model trained via meta-learning enables fine-tuning to adapt solutions for subsidiary problems once the initial training is complete. Additionally, we enhance the PPO-CLIP framework with a novel strategy that integrates probability shifts and Generalized Advantage Estimation (GAE). In the final stage of scheduling plan selection, a technique for identifying interval turning points is employed to choose the optimal plan from the Pareto solution set. The results demonstrate that the method not only secures excellent scheduling solutions in complex environments through its robust generalization capabilities but also shows significant improvements over interval-constrained multi-objective evolutionary algorithms, such as IP-MOEA, ICMOABC, and IMOMA-II, across multiple multi-objective evaluation metrics including hypervolume (HV), runtime, and uncertainty.

High‐order multilayer attention fusion network for 3D object detection

AbstractThree‐dimensional object detection based on the fusion of 2D image data and 3D point clouds has become a research hotspot in the field of 3D scene understanding. However, different sensor data have discrepancies in spatial position, scale, and alignment, which severely impact detection performance. Inappropriate fusion methods can lead to the loss and interference of valuable information. Therefore, we propose the High‐Order Multi‐Level Attention Fusion Network (HMAF‐Net), which takes camera images and voxelized point clouds as inputs for 3D object detection. To enhance the expressive power between different modality features, we introduce a high‐order feature fusion module that performs multi‐level convolution operations on the element‐wise summed features. By incorporating filtering and non‐linear activation, we extract deep semantic information from the fused multi‐modal features. To maximize the effectiveness of the fused salient feature information, we introduce an attention mechanism that dynamically evaluates the importance of pooled features at each level, enabling adaptive weighted fusion of significant and secondary features. To validate the effectiveness of HMAF‐Net, we conduct experiments on the KITTI dataset. In the “Car,” “Pedestrian,” and “Cyclist” categories, HMAF‐Net achieves mAP performances of 81.78%, 60.09%, and 63.91%, respectively, demonstrating more stable performance compared to other multi‐modal methods. Furthermore, we further evaluate the framework's effectiveness and generalization capability through the KITTI benchmark test, and compare its performance with other published detection methods on the 3D detection benchmark and BEV detection benchmark for the “Car” category, showing excellent results. The code and model will be made available on https://github.com/baowenzhang/High‐order‐Multilayer‐Attention‐Fusion‐Network‐for‐3D‐Object‐Detection.

Deep-TCP: Multi-source data fusion for deep learning-powered tropical cyclone intensity prediction to enhance urban sustainability

Tropical cyclones (TC) exert a profound impact on cities, causing extensive damage and losses. Thus, TC Intensity Prediction is crucial for creating sustainable cities as it enables proactive measures to be taken, including evacuation planning, infrastructure reinforcement, and emergency response coordination. In this study, we propose a Deep learning-powered TC Intensity Prediction (Deep-TCP) framework. In particular, Deep-TCP contains a data constraint module for fusing data features from multiple sources and establishing a unified global representation. To capture the spatiotemporal attributes, a Spatial-Temporal Attention (ST-Attention) module is built to distill insights from environmental variables. To improve the robustness and stability of the predictions, an encoder-decoder module that utilizes the ConvGPU unit is introduced to enhance feature maps. Then, a novel feature enhancement module is built to bolster the generalization capability and solve the dependency attenuation. The results demonstrate that the Deep-TCP framework significantly outperforms various benchmarks. Additionally, it effectively predicts multiple TC categories within the 6–24 h timeframe, showing strong capability in predicting changing trends. The reliable prediction results are potentially beneficial for disaster management and urban planning, significantly enhancing urban sustainability by improving preparedness and response strategies.

Volume fraction detection in multiphase systems using neutron activation analysis and artificial neural network

This study presents an application of an Artificial Neural Network (ANN) to detect fluids in an annular flow regime using Prompt-Gamma Neutron Activation Analysis (PGNAA). The ANN was trained using gamma-ray spectra resulting from neutron interactions with chemical elements found in fluids typical of multiphase flow in oil exploration. These spectra were generated through mathematical simulation using the MCNP6 Monte Carlo computer code to model nuclear particle transport. A241Am-Be polyenergetic neutron source was simulated for these calculations. Several combinations of fluid fractions were developed to create a dataset used for both training and evaluation of the ANN. The ANN demonstrated robust generalization capabilities by accurately predicting the volume fraction of the three investigated fluids (saltwater, oil, and gas), even for cases not included in the training phase. The combination of ANN and PGNAA proved effective for analyzing multiphase systems, with over 92% of all showing errors of less than 5%.