Search Results (27)

Advanced deception countermeasures now enable adversaries to inject false targets into synthetic-aperture-radar (SAR) imagery, generating electromagnetic signatures virtually indistinguishable from genuine targets, thus destroying the separability essential for conventional recognition algorithms. To address this problem, we propose a versatile front-end Multi-Feature Extraction and Spatial Transformation Network (MFE-STN), specifically designed for the task of discriminating between true targets and deceptive false targets created by SAR jamming, which can be seamlessly integrated with existing CNN backbones without architecture modification. MFE-STN integrates three complementary operations: (i) wavelet decomposition to extract the overall geometric features and scattering distribution of the target, (ii) a manifold transformation module for non-linear alignment of heterogeneous feature spaces, and (iii) a lightweight deformable spatial transformer that compensates for local geometric distortions introduced by deceptive jamming. By analyzing seven typical parameter-mismatch effects, we construct a simulated dataset containing six representative classes—four known classes and two unseen classes. Experimental results demonstrate that inserting MFE-STN boosts the average F1-score of known targets by 12.19% and significantly improves identification accuracy for unseen targets. This confirms the module’s capability to capture discriminative signatures to distinguish genuine targets from deceptive ones while exhibiting strong cross-domain generalization capabilities. Full article

Modern credit scoring systems must operate under increasingly complex borrower data conditions, characterized by structural heterogeneity and regulatory demands for transparency. This study proposes a modular modeling framework that addresses both interpretability and data incompleteness in credit risk prediction. By leveraging Weight of Evidence (WoE) binning and logistic regression, we constructed domain-specific sub-models that correspond to different attribute sets and integrated them through ensemble, hierarchical, and stacking-based architectures. Using a real-world dataset from the American Express default prediction challenge, we demonstrate that these modular architectures maintain high predictive performance (test Gini > 0.90) while preserving model transparency. Comparative analysis across multiple architectural designs highlights trade-offs between generalization, computational complexity, and regulatory compliance. Our main contribution is a systematic comparison of logistic regression–based architectures that balances accuracy, robustness, and interpretability. These findings highlight the value of modular decomposition and stacking for building predictive yet interpretable credit risk models. Full article

The present study focuses on wind power forecasting (WPF) models based on deep neural networks (DNNs), aiming to evaluate current approaches, identify gaps, and provide insights into their importance for the integration of Renewable Energy Sources (RESs). The systematic review was conducted following the methodology of Kitchenham and Charters, including peer-reviewed articles from 2020 to 2024 that focused on WPF using deep learning (DL) techniques. Searches were conducted in the ACM Digital Library, IEEE Xplore, ScienceDirect, Springer Link, and Wiley Online Library, with the last search updated in April 2024. After the first phase of screening and then filtering using inclusion and exclusion criteria, risk of bias was assessed using a Likert-scale evaluation of methodological quality, validity, and reporting. Data extraction was performed for 120 studies. The synthesis established that the state of the art is dominated by hybrid architectures (e.g., CNN-LSTM) integrated with signal decomposition techniques like VMD and optimization algorithms such as GWO and PSO, demonstrating high predictive accuracy for short-term horizons. Despite these advancements, limitations include the variability in datasets, the heterogeneity of model architectures, and a lack of standardization in performance metrics, which complicate direct comparisons across studies. Overall, WPF models based on DNNs demonstrate substantial promise for renewable energy integration, though future work should prioritize standardization and reproducibility. This review received no external funding and was not prospectively registered. Full article

Financial markets exhibit fundamental asymmetries in temporal causality, where policy interventions create asymmetric transmission patterns that traditional symmetric modeling approaches fail to capture. This work introduces a mathematical framework that exploits the inherent symmetries of transformer architectures while preserving essential asymmetric temporal relationships in financial causal inference. We develop CausalFormer, a symmetry-aware neural architecture that maintains the permutation equivariance properties of self-attention mechanisms while enforcing strict temporal asymmetry constraints for causal discovery. The framework incorporates three mathematically principled components: (1) a symmetric attention matrix construction with asymmetric temporal masking that preserves the mathematical elegance of transformer operations while ensuring causal consistency, (2) a multi-scale convolution module with symmetric kernel initialization but asymmetric temporal receptive fields that captures policy transmission effects across heterogeneous time horizons, and (3) enhanced Nelson–Siegel decomposition that maintains the symmetric factor structure while modeling the evolution dynamics of asymmetric factors. Our mathematical formulation establishes the formal symmetry properties of the attention mechanism under temporal transformations while proving asymmetric convergence behaviors in policy transmission scenarios. The integration of symmetric optimization landscapes with asymmetric causal constraints enables simultaneous achievement of mathematical elegance and economic interpretability. Comprehensive experiments on monetary policy datasets demonstrate that the symmetry-aware design achieves a 15.3% improvement in the accuracy of causal effect estimations and a 12.7% enhancement in the predictive performance compared to those for existing methods while maintaining 91.2% causal consistency scores. The framework successfully identifies asymmetric policy transmission mechanisms, revealing that monetary tightening exhibits 40% faster propagation than easing policies, establishing new mathematical insights into the temporal asymmetries in financial systems. This work demonstrates how principled exploitation of architectural symmetries combined with domain-specific asymmetric constraints opens up new directions for mathematically rigorous yet economically interpretable deep learning in financial econometrics, with broad applications spanning computational finance, economic forecasting, and policy analysis. Full article

The successful integration of Terrestrial and Non-Terrestrial Networks (T/NTNs) in 6G is poised to revolutionize demanding domains like Earth Observation (EO) and Intelligent Transportation Systems (ITSs). Still, it requires Distributed Machine Learning (DML) frameworks that are scalable, private, and efficient. Existing methods, such as Federated Learning (FL) and Split Learning (SL), face critical limitations in terms of client computation burden and latency. To address these challenges, this paper proposes a novel hierarchical DML paradigm. We first introduce Federated Split Transfer Learning (FSTL), a foundational framework that synergizes FL, SL, and Transfer Learning (TL) to enable efficient, privacy-preserving learning within a single client group. We then extend this concept to the Generalized FSTL (GFSTL) framework, a scalable, multi-group architecture designed for complex and large-scale networks. GFSTL orchestrates parallel training across multiple client groups managed by intermediate servers (RSUs/HAPs) and aggregates them at a higher-level central server, significantly enhancing performance. We apply this framework to a unified T/NTN architecture that seamlessly integrates vehicular, aerial, and satellite assets, enabling advanced applications in 6G ITS and EO. Comprehensive simulations using the YOLOv5 model on the Cityscapes dataset validate our approach. The results show that GFSTL not only achieves faster convergence and higher detection accuracy but also substantially reduces communication overhead compared to baseline FL, and critically, both detection accuracy and end-to-end latency remain essentially invariant as the number of participating users grows, making GFSTL especially well suited for large-scale heterogeneous 6G ITS deployments. We also provide a formal latency decomposition and analysis that explains this scaling behavior. This work establishes GFSTL as a robust and practical solution for enabling the intelligent, connected, and resilient ecosystems required for next-generation transportation and environmental monitoring. Full article

Intelligent edge machine learning has emerged as a paradigm for deploying smart applications across resource-constrained devices in next-generation network infrastructures. This survey addresses the critical challenges of implementing machine learning models on edge devices within distributed network environments, including computational limitations, memory constraints, and energy-efficiency requirements for real-time intelligent inference. We provide comprehensive analysis of soft computing optimization strategies essential for intelligent edge deployment, systematically examining model compression techniques including pruning, quantization methods, knowledge distillation, and low-rank decomposition approaches. The survey explores intelligent MLOps frameworks tailored for network edge environments, addressing continuous model adaptation, monitoring under data drift, and federated learning for distributed intelligence while preserving privacy in next-generation networks. Our work covers practical applications across intelligent smart agriculture, energy management, healthcare, and industrial monitoring within network infrastructures, highlighting domain-specific challenges and emerging solutions. We analyze specialized hardware architectures, cloud offloading strategies, and distributed learning approaches that enable intelligent edge computing in heterogeneous network environments. The survey identifies critical research gaps in multimodal model deployment, streaming learning under concept drift, and integration of soft computing techniques with intelligent edge orchestration frameworks for network applications. These gaps directly manifest as open challenges in balancing computational efficiency with model robustness due to limited multimodal optimization techniques, developing sustainable intelligent edge AI systems arising from inadequate streaming learning adaptation, and creating adaptive network applications for dynamic environments resulting from insufficient soft computing integration. This comprehensive roadmap synthesizes current intelligent edge machine learning solutions with emerging soft computing approaches, providing researchers and practitioners with insights for developing next-generation intelligent edge computing systems that leverage machine learning capabilities in distributed network infrastructures. Full article

Karst springs play a critical strategic role in regional economic and ecological sustainability, yet their spatiotemporal heterogeneity and hydrological complexity pose substantial challenges for flow prediction. This study proposes FMD-mGTO-BiGRU-KAN, a four-stage hybrid deep learning architecture for daily spring flow prediction that integrates multi-feature signal decomposition, meta-heuristic optimization, and interpretable neural network design: constructing an Feature Mode Decomposition (FMD) decomposition layer to mitigate modal aliasing in meteorological signals; employing the improved Gorilla Troops Optimizer (mGTO) optimization algorithm to enable autonomous hyperparameter evolution, overcoming the limitations of traditional grid search; designing a Bidirectional Gated Recurrent Unit (BiGRU) network to capture long-term historical dependencies in spring flow sequences through bidirectional recurrent mechanisms; introducing Kolmogorov–Arnold Networks (KAN) to replace the fully connected layer, and improving the model interpretability through differentiable symbolic operations; Additionally, residual modules and dropout blocks are incorporated to enhance generalization capability, reduce overfitting risks. By integrating multiple deep learning algorithms, this hybrid model leverages their respective strengths to adeptly accommodate intricate meteorological conditions, thereby enhancing its capacity to discern the underlying patterns within complex and dynamic input features. Comparative results against benchmark models (LSTM, GRU, and Transformer) show that the proposed framework achieves 82.47% and 50.15% reductions in MSE and RMSE, respectively, with the NSE increasing by 8.01% to 0.9862. The prediction errors are more tightly distributed, and the proposed model surpasses the benchmark model in overall performance, validating its superiority. The model’s exceptional prediction ability offers a novel high-precision solution for spring flow prediction in complex hydrological systems. Full article

To improve the traffic efficiency of connected and automated vehicles (CAVs) in on-ramp merging areas, this study proposes a novel region-level multi-agent reinforcement learning framework, Regionally Coordinated Nash-Advantage Decomposition Deep Q-Network with Conflict-Aware Q Fusion (RC-NashAD-DQN). Unlike existing vehicle-level control methods, which suffer from high computational overhead and poor scalability, our approach abstracts on-ramp and main road areas as region-level control agents, achieving coordinated yet independent decision-making while maintaining control precision and merging efficiency comparable to fine-grained vehicle-level approaches. Each agent adopts a value–advantage decomposition architecture to enhance policy stability and distinguish action values, while sharing state–action information to improve inter-agent awareness. A Nash equilibrium solver is applied to derive joint strategies, and a conflict-aware Q-fusion mechanism is introduced as a regularization term rather than a direct action-selection tool, enabling the system to resolve local conflicts—particularly at region boundaries—without compromising global coordination. This design reduces training complexity, accelerates convergence, and improves robustness against communication imperfections. The framework is evaluated using the SUMO simulator at the Taishan Road interchange on the S1 Yongtaiwen Expressway under heterogeneous traffic conditions involving both passenger cars and container trucks, and is compared with baseline models including C-DRL-VSL and MADDPG. Extensive simulations demonstrate that RC-NashAD-DQN significantly improves average traffic speed by 17.07% and reduces average delay by 12.68 s, outperforming all baselines in efficiency metrics while maintaining robust convergence performance. These improvements enhance cooperation and merging efficiency among vehicles, contributing to sustainable urban mobility and the advancement of intelligent transportation systems. Full article

The critical component of the vision transformer (ViT) architecture is multi-head self-attention (MSA), which enables the encoding of long-range dependencies and heterogeneous interactions. However, MSA has two significant limitations: its limited ability to capture local features and its high computational costs. To address these challenges, this paper proposes an integrated multi-head self-attention approach with a bottleneck enhancement structure, named WMSA–WBS, which mitigates the aforementioned shortcomings of conventional MSA. Different from existing wavelet-enhanced ViT variants that mainly focus on the isolated wavelet decomposition in the attention layer, WMSA–WBS introduces a co-design of wavelet-based frequency processing and bottleneck optimization, achieving more efficient and comprehensive feature learning. Within WMSA–WBS, the proposed wavelet multi-head self-attention (WMSA) approach is combined with a novel wavelet bottleneck structure to capture both global and local information across the spatial, frequency, and channel domains. Specifically, this module achieves these capabilities while maintaining low computational complexity and memory consumption. Extensive experiments demonstrate that ViT models equipped with WMSA–WBS achieve superior trade-offs between accuracy and model complexity across various vision tasks, including image classification, object detection, and semantic segmentation. Full article

Ultra-short-term PV power prediction (USTPVPP) results provide a basis for the development of intra-day rolling power generation plans. However, due to the feature information and the unpredictability of meteorology, the current ultra-short-term PV power prediction accuracy improvement still faces technical challenges. In this paper, we propose a combined prediction framework that takes into account the reconfiguration of the predictable components of PV stations and the spatiotemporal heterogeneous maps. A circuit singular spectral decomposition (CISSD) intrinsic predictable component extraction method is adopted to obtain specific frequency components in sensitive meteorological variables, a mechanism based on radiation characteristics and PV power trend predictable component extraction and reconstruction is proposed to enhance power predictability, and a spatiotemporal heterogeneous graph neural network (STHGNN) combined with a Non-stationary Transformer (Ns-Transformer) combination architecture to achieve joint prediction for different PV components. The proposed method is applied to a PV power plant in Gansu, China, and the results show that the prediction method based on the proposed combined spatio-temporal heterogeneous graph neural network model combined with the proposed predictable component extraction achieves an average reduction of 6.50% in the RMSE, an average reduction of 2.50% in the MAE, and an average improvement of 11.93% in the R² over the direct prediction method, respectively. Full article

Wind power forecasting is challenging because of complex, nonlinear relationships between inherent patterns and external disturbances. Though much progress has been achieved in deep learning approaches, existing methods cannot effectively decompose and model intertwined spatio-temporal dependencies. Current methods typically treat wind power as a unified signal without explicitly separating inherent patterns from external influences, so they have limited prediction accuracy. This paper introduces a novel framework GCN-EIF that decouples external interference factors (EIFs) from inherent wind power patterns to achieve excellent prediction accuracy. Our innovation lies in the physically informed architecture that explicitly models the mathematical relationship: $P\left(t\right)=P_{\mathrm{inherent}}\left(t\right)+\mathrm{EIF}\left(t\right)$. The framework adopts a three-component architecture consisting of (1) a multi-graph convolutional network using both geographical proximity and power correlation graphs to capture heterogeneous spatial dependencies between wind farms, (2) an attention-enhanced LSTM network that weights temporal features differentially based on their predictive significance, and (3) a specialized Conv2D mechanism to identify and isolate external disturbance patterns. A key methodological contribution is our signal decomposition strategy during the prediction phase, where an EIF is eliminated from historical data to better learn fundamental patterns, and then a predicted EIF is reintroduced for the target period, significantly reducing error propagation. Extensive experiments across diverse wind farm clusters and different weather conditions indicate that GCN-EIF achieves an 18.99% lower RMSE and 5.08% lower MAE than state-of-the-art methods. Meanwhile, real-time performance analysis confirms the model’s operational viability as it maintains excellent prediction accuracy (RMSE < 15) even at high data arrival rates (100 samples/second) while ensuring processing latency below critical thresholds (10 ms) under typical system loads. Full article

This paper introduces Harmonizer, a universal framework designed for tokenizing heterogeneous input signals, including text, audio, and video, to enable seamless integration into multimodal large language models (LLMs). Harmonizer employs a unified approach to convert diverse, non-linguistic signals into discrete tokens via its FusionQuantizer architecture, built on FluxFormer, to efficiently capture essential signal features while minimizing complexity. We enhance features through STFT-based spectral decomposition, Hilbert transform analytic signal extraction, and SCLAHE spectrogram contrast optimization, and train using a composite loss function to produce reliable embeddings and construct a robust vector vocabulary. Experimental validation on music datasets such as E-GMD v1.0.0, Maestro v3.0.0, and GTZAN demonstrates high fidelity across 288 s of vocal signals (MSE = 0.0037, CC = 0.9282, Cosine Sim. = 0.9278, DTW = 12.12, MFCC Sim. = 0.9997, Spectral Conv. = 0.2485). Preliminary tests on text reconstruction and UCF-101 video clips further confirm Harmonizer’s applicability across discrete and spatiotemporal modalities. Rooted in the universality of wave phenomena and Fourier theory, Harmonizer offers a physics-inspired, modality-agnostic fusion mechanism via wave superposition and interference principles. In summary, Harmonizer integrates natural language processing and signal processing into a coherent tokenization paradigm for efficient, interpretable multimodal learning. Full article

As global energy demand continues to rise, accurate load forecasting has become increasingly crucial for power system operations. This study proposes a novel Complete Ensemble Empirical Mode Decomposition with Adaptive Noise-Fast Fourier Transform-inverted Transformer-Long Short-Term Memory (CEEMDAN-FFT-iTransformer-LSTM) methodological framework to address the challenges of component complexity and transient fluctuations in power load sequences. The framework initiates with CEEMDAN-based signal decomposition, which dissects the original load sequence into multiple intrinsic mode functions (IMFs) characterized by different temporal scales and frequencies, enabling differentiated processing of heterogeneous signal components. A subsequent application of Fast Fourier Transform (FFT) extracts discriminative frequency-domain features, thereby enriching the feature space with spectral information. The architecture employs an iTransformer module with multi-head self-attention mechanisms to capture high-frequency patterns in the most volatile IMFs, while a gated recurrent unit (LSTM) specializes in modeling low-frequency components with longer temporal dependencies. Experimental results demonstrate the proposed framework achieves superior performance with an average 80% improvement in R-squared ($R^2$), 40.1% lower Mean Absolute Error (MAE), and 54.1% reduced Mean Squared Error (RMSE) compared to other models. This advancement provides a robust computational tool for power grid operators, enabling optimal resource dispatch through enhanced prediction accuracy to reduce operational costs. The demonstrated capability to resolve multi-scale temporal dynamics suggests potential extensions to other forecasting tasks in energy systems involving complex temporal patterns. Full article

The forecasting of high-water-cut oil well production faces challenges of strong nonlinearity and nonstationarity due to reservoir heterogeneity and multiscale dynamic characteristics. This study proposes a hybrid CEEMDAN-SR-BiLSTM framework based on a “decomposition-feature enhancement-integration” architecture. The framework employs Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) to suppress mode mixing, reconstructs high-, medium-, and low-frequency subsequences using Hilbert-Huang Transform (HHT) combined with tercile thresholding, and finally achieves multiscale feature fusion prediction through a Bayesian-optimized bidirectional long short-term memory network (BiLSTM). Interpretability analysis based on SHapley Additive exPlanations (SHAP) values reveals the contribution degrees of parameters such as water injection volume and flowing pressure to different frequency components, establishing a mapping between production data features and physical mechanisms of oil well production. This mapping, integrated with physical mechanisms including wellbore transient flow, injection-production response lag, and reservoir pressure evolution, enables mechanistic interpretation of production phenomena and quantitative decoupling and prediction of multiscale dynamics. Experimental results show that the framework achieves a root-mean-square error (RMSE) of 3.75 in forecasting a high-water-cut well (water cut = 87.6%) in the Qaidam Basin, reducing errors by 26.0% and 50.0% compared to CEEMDAN-BiLSTM and BiLSTM models, respectively, with a coefficient of determination (R²) reaching 0.954. Full article

Elastic wavefield separation in anisotropic media is essential for seismic imaging but remains challenging due to complex interactions among multiple wave modes. Traditional methods often rely on solving the Christoffel equation, which is computationally expensive, particularly in heterogeneous models. This study proposes a deep learning-based approach using a cross-attention U-Net architecture to achieve efficient vector decomposition of elastic wavefields. The model employs a dual-branch encoder with cross-attention mechanisms to preserve and exploit inter-component relationships among wavefield components. The network was trained on patches extracted from the BP (British Petroleum) 2007 anisotropic benchmark model, with ground truth labels being generated via low-rank approximation methods. Quantitative evaluations show that the cross-attention U-Net outperforms a baseline U-Net, improving the peak signal-to-noise ratio(PSNR) by 1.25 dB (44.10 dB vs. 42.85 dB) and structural similarity index (SSIM) by 0.014 (0.904 vs. 0.890). The model demonstrates effective generalization to larger domains and different geological settings, validated on both the extended BP model and the Hess vertically transversely isotropic (VTI) model. Overall, this approach provides a computationally efficient alternative to traditional separation methods while maintaining physical consistency in the separated wavefields. Full article