Search Results (94)

This study evaluates a hierarchical workflow for fault diagnosis in transmission systems by integrating established machine-learning techniques for fault-type classification and zonal fault location with a complementary double-ended traveling-wave procedure for point-location estimation. The contribution lies at the level of process integration and comparative evaluation rather than in the proposal of a new machine-learning or traveling-wave algorithm. The methodology was evaluated using the IEEE 9-bus test system. Symmetrical and asymmetrical short-circuit scenarios were automatically simulated at multiple positions along six transmission lines using DIgSILENT PowerFactory, and the resulting oscillographic records were exported in COMTRADE format, producing a database of 2952 fault events. Phase voltages and currents, together with positive-, negative-, and zero-sequence components, were used to evaluate Decision Trees, Self-Organizing Maps (SOM), Artificial Neural Networks (ANN), and k-Nearest Neighbors (KNN) for fault-type classification and zonal fault location. Under the simulated noise-free conditions and the adopted fixed hold-out partition, all four algorithms correctly classified the 591 fault-type testing observations, yielding 100% test-set accuracy. This result characterizes the specific evaluation subset considered in the study; repeated, cross-validated, or grouped partitions were not performed, and neighboring simulated fault positions may therefore be represented across the training and testing subsets. For zonal fault location, the ANN exhibited the strongest and most consistent observed performance in the retained 590-event evaluation set, with class-specific recall values between approximately 0.96 and 0.99 across the six fault zones, whereas the Decision Tree provided a favorable compromise between zonal discrimination and computational efficiency. Some model-specific hyperparameter values from the original executions are unavailable in the retained experimental record, which limits exact replication of those original configurations; the reported results correspond to the evaluated executions documented in this study. As a complementary third component of the workflow, the double-ended traveling-wave procedure based on discrete wavelet analysis was illustrated for one AG event simulated at 25% of the transmission-line length, producing a normalized point-location estimate of approximately 25.07% from the local terminal. This single-event analysis demonstrates the operation of the traveling-wave processing sequence, while broader multi-event validation is outside the present experimental scope. Overall, the results demonstrate the coordinated application of fault-type classification, zonal fault location, and traveling-wave-based point-location refinement within a common diagnostic workflow. The findings should be interpreted within the deterministic simulation conditions, fixed evaluation subsets, and experimental records considered in this study. Full article

The electromagnetic spectrum grows increasingly crowded with the rapid expansion of mobile, satellite and Internet of Things communications, making intelligent spectrum sensing and efficient management an urgent priority. Automatic modulation classification (AMC) serves as the core of cognitive radio and intelligent communication. Existing deep models often suffer from a large number of parameters and low storage efficiency. To overcome these limitations, we propose MDSCNet, a multi-scale depth-wise separable complex network. Built upon complex depth-wise separable convolution, the network makes full use of the phase information in in-phase and quadrature signals while naturally preserving the symmetric relationship between the in-phase and quadrature components (IQ). The asymmetric multi-scale structure combined with the embedded lightweight attention module jointly forms the overall feature extraction process. The overall parameter count is kept extremely low, at only 47.739 k. Experiments on the RML2016.10a and RML2016.10b datasets show that MDSCNet delivers recognition performance under low signal-to-noise ratios (SNR), reaching 63.42% and 66.71% respectively. More importantly, it outperforms mainstream methods in both parameter count and storage efficiency. Full article

Gradient-boosted trees outperform neural networks on tabular data without privacy, often taken to imply that differentially private federated learning should be based on boosting. We revisit this implication under matched accounting—a single privacy-loss distribution accountant, cross-checked against Rényi accounting—and symmetric, per-budget tuning, and find limited support for this expectation. On the Diabetes 130-US-Hospitals and BRFSS datasets across ε ∈ {0.5, 1, 2, 4, 8} over 20 seeds, differentially private federated boosting, a differentially private stochastic gradient descent (DP-SGD) network, and DP-SGD logistic regression achieve similar performance; no model class is consistently superior. On the real corpora, the most frugal model wins at the tightest budget—logistic regression is best at ε = 0.5 (0.589 and 0.811 AUC)—while the network leads slightly at looser budgets; boosting remains competitive but does not lead. On the synthetic task, boosting leads at tight budgets and the network at looser ones. The comparison is asymmetrically tuning-sensitive: fixing the boosting round count can produce an apparent neural advantage, whereas the network is robust to its step count. Off-path privatization and sequential noise accumulation explain the behavior; boosting’s main advantage is not accuracy but communication, achieving one to three orders of magnitude fewer values per client. Full article

Asymmetric interaction matrices encode the linear coupling structure and directed interaction patterns that arise in mathematical models of complex networks across ecology, finance, and machine learning, yet their geometric structure as points on a statistical manifold has received comparatively little systematic treatment. This paper develops a rigorous information-geometric framework for the manifold ${\mathcal{M}}_n^{+}$ of real $n\times n$ interaction matrices whose symmetric part is negative definite—equivalently, the matrices satisfying the numerical stability condition $\omega \left(A\right)={\lambda }_{\max }\left(S\left(A\right)\right)<0$. The symmetric part $S\left(A\right)=(A+A^{\mathsf{T}})/2$ and the skew-symmetric part $K\left(A\right)=(A-A^{\mathsf{T}})/2$ correspond, respectively, to the metric structure and the torsion of the induced statistical manifold. We construct the natural augmented Riemannian metric g on ${\mathcal{M}}_n^{+}$ as the sum of the Fisher–Rao pullback metric through $-S(·)$ and a Frobenius term on $K(·)$, derive explicit formulae for the sectional curvature in the mixed symmetric–skew plane, and prove that the sectional curvature vanishes if and only if A is normal. The central theoretical result is a curvature-mediated stability theorem: a Fisher–Rao stability margin, derived from the precision representative of A, provides a sharp, computationally accessible certificate for the asymptotic stability of the linear dynamical system $\dot{\mathbf{x}}=A\mathbf{x}$, with the instability boundary lying at infinite Fisher–Rao distance. We further establish an information-geometric reformulation of May’s stability criterion for random ecological networks, a curvature-based covariance regularisation scheme for financial correlation matrices, and a Jacobian stability bound for deep neural networks. All the main results are illustrated with explicit $3\times 3$ and $4\times 4$ numerical examples. Full article

Symmetrical resistance training induces corticospinal and cortical network plasticity; however, the neural consequences of asymmetrical resistance training (offset loading or OL) remain unclear. In this preliminary study, fourteen healthy adults completed eighteen supervised upper-extremity training sessions over a 6-week intervention period. Participants were allocated to either an offset loading (OL) intervention group (n = 8) or a conventional symmetrical resistance training active control group (AC; n = 6). Electroencephalography (EEG) recordings were acquired during motor task performance at baseline, 3 weeks, and 6 weeks. Directed functional connectivity among predefined cortical regions was quantified, and longitudinal changes were assessed using Friedman tests with false-discovery-rate correction for multiple comparisons. Significant changes in the OL group were observed involving two cortical pathways after correction for multiple testing. Connectivity from the right parietal cortex to the right sensorimotor cortex increased over time (Friedman χ² = 8.00, q = 0.037), with post hoc analyses showing a significant increase between the midpoint and post-training assessments (q = 0.047; effect size r = 0.894). No significant longitudinal changes in cortical connectivity were identified in the AC group. Six weeks of OL training was associated with selective strengthening of directed connectivity within prefrontal-to-sensorimotor and parietal-to-sensorimotor cortical pathways. In contrast, conventional symmetrical resistance training was not associated with detectable changes in connectivity. These findings suggest that asymmetrical loading may induce task-specific reorganization of cortical networks involved in sensorimotor processing and motor control. Full article

Byzantine failures remain a critical threat to Federated Learning (FL), where malicious clients inject adversarial updates to disrupt global model convergence. From the perspective of symmetry, benign client updates typically exhibit statistical symmetry around the global consensus, whereas Byzantine attacks function as “symmetry-breaking” events that introduce skewness and distributional anomalies. Existing defenses often rely on unrealistic assumptions or fail to capture these asymmetric deviations under high-dimensional non-IID settings. In this paper, we propose a symmetry-aware Byzantine-resilient FL framework driven by a Dual-Channel Attention-Driven Anomaly Detector (DAAD). Specifically, DAAD transforms inter-client behaviors into geometrically symmetric interaction matrices—encoding Gradient Cosine Similarities and Loss Euclidean Distances—to construct dual-channel spatial representations. These representations are processed via a Convolutional Neural Network (CNN) enhanced with Squeeze-and-Excitation (SE) attention blocks, which leverage the inherent symmetry of benign consensus to extract robust adversarial signatures. The detector is pre-trained offline on a synthetic dataset incorporating a diverse portfolio of simulated attacks (e.g., Gaussian noise and label flipping). Crucially, this pre-trained model is seamlessly embedded into the online FL loop to filter updates without requiring ground-truth labels. By jointly encoding client behaviors and learning cross-modal attack signatures, our framework enables reliable detection even when over half of the clients are Byzantine. Extensive experiments on MNIST, CIFAR-10, and FEMNIST datasets demonstrate that DAAD consistently outperforms existing robust aggregation baselines in both anomaly detection accuracy and global model performance, especially under high Byzantine ratios and non-IID conditions. Full article

Traffic flow prediction is a pivotal task in intelligent transportation systems. The primary challenge lies in accurately modeling the dynamically evolving and directional spatio-temporal dependencies inherent in road networks. Existing graph neural network-based methods suffer from three main limitations: (1) symmetric adjacency matrices fail to capture the causal propagation of traffic flow from upstream to downstream; (2) the serial combination of graph and temporal convolutions lacks an explicit modeling of joint spatio-temporal state transition probabilities; (3) the inherent low-pass filtering property of temporal convolutional networks tends to smooth high-frequency abrupt signals, thereby weakening responsiveness to sudden events. To address these issues, this paper proposes a causal-enhanced spatio-temporal Markov graph convolutional network (CSHGCN). At the spatial modeling level, we construct an asymmetric causal adjacency matrix by decoupling source and target node embeddings to learn directional traffic flow influences. At the spatio-temporal joint modeling level, we design a spatio-temporal Markov transition module (STMTM) based on spatio-temporal Markov chain theory, which explicitly learns conditional transition patterns through temporal dependency encoders, spatial dependency encoders, and a joint transition network. At the temporal modeling level, we introduce differential feature enhancement and high-frequency residual compensation mechanisms to preserve key abrupt change information through frequency-domain complementarity. Experiments on four datasets—PEMS03, PEMS04, PEMS07, and PEMS08—demonstrate that CSHGCN outperforms existing baselines in terms of MAE, RMSE, and MAPE, with ablation studies validating the effectiveness of each module. Full article

In modern power distribution networks, robust and intelligent fault management techniques are increasingly important as system complexity grows with the integration of distributed energy resources. This article reviews the use of deep learning methods for short-circuit fault detection, classification, and localization in power distribution systems, including symmetrical, asymmetrical, and high-impedance faults. The approaches examined include convolutional neural networks, recurrent neural networks, deep reinforcement learning, graph neural networks, and hybrid architectures. A comprehensive taxonomy of these models is presented, followed by an analysis of their application across the stages of fault diagnostics. Major contributions to the field are highlighted, and research gaps are identified in relation to data scarcity, model interpretability, real-time responsiveness, and deployment scalability. The paper provides an in-depth technical and performance comparison of deep learning approaches based on current research trends, and it also outlines the limitations of previous review studies. The objective of this work is to support researchers in selecting and implementing appropriate deep learning techniques for fault analytics in complex smart electricity grids with significant penetration of distributed energy resources. The review is intended to serve as an initial foundation for continued research and development in intelligent fault analytics for reliable and sustainable power distribution systems. Full article

Symmetry and asymmetry are foundational design principles in artificial intelligence, defining the balance between invariance and adaptability in multimodal learning systems. In audio-visual speaker verification, where speech and lip-motion features are jointly modeled to determine whether two utterances belong to the same individual, these principles govern both fairness and discriminative power. In this work, we analyze how symmetry and asymmetry emerge within a gated-fusion architecture that integrates Time-Delay Neural Networks and Bidirectional Long Short-Term Memory encoders for speech, ResNet-based visual lip encoders, and a shared Conformer-based temporal backbone. Structural symmetry is preserved through weight-sharing across paired utterances and symmetric cosine-based scoring, ensuring verification consistency regardless of input order. In contrast, asymmetry is intentionally introduced through modality-dependent temporal encoding, multi-head attention pooling, and a learnable gating mechanism that dynamically re-weights the contribution of audio and visual streams at each timestep. This controlled asymmetry allows the model to rely on visual cues when speech is noisy, and conversely on speech when lip visibility is degraded, yielding adaptive robustness under cross-modal degradation. Experimental results demonstrate that combining symmetric embedding space design with adaptive asymmetric fusion significantly improves generalization, reducing Equal Error Rate (EER) to 3.419% on VoxCeleb-2 test dataset without sacrificing interpretability. The findings show that symmetry ensures stable and fair decision-making, while learnable asymmetry enables modality awareness together forming a principled foundation for next-generation audio-visual speaker verification systems. Full article

Image inpainting, a pivotal technology for restoring damaged regions of images, has emerged as a significant research focus in computer vision. This review systematically surveys recent advances in deep learning-based image inpainting. We begin by categorizing prevailing methods into three groups based on their core architectures: Convolutional Neural Networks (CNNs), Generative Models, and Transformers. Through a comparative analysis of their symmetric versus asymmetric network architectures, applicable scenarios, and performance bottlenecks, we provide a critical discussion of the strengths and limitations inherent to each approach. The evolution of underlying design principles, such as symmetry, and the corresponding solutions to core challenges are also discussed. Furthermore, we introduce key benchmark datasets and commonly used image quality assessment metrics, offering a multidimensional framework for evaluation. We highlight that mainstream datasets collectively foster a greenhouse-like evaluation environment detached from real-world complexities and that existing metrics are critically misaligned with the fundamental objective of inpainting: generating plausible new content. Finally, we summarize the prevailing challenges in current deep learning-based inpainting research and outline promising future directions. We highlight critical issues, such as enhancing restoration quality, reducing computational costs, and broadening application scenarios, thereby providing valuable insights for subsequent research. Full article

Microservice architecture has become a foundational component of modern distributed systems due to its modularity, scalability, and deployment flexibility. However, the increasing complexity and dynamic nature of service interactions have introduced substantial challenges in accurately detecting runtime anomalies. Existing methods often rely on multiple monitoring metrics, which introduce redundancy and noise while increasing the complexity of data collection and model design. This paper proposes a novel spatiotemporal anomaly detection framework that integrates Dynamic Graph Neural Networks (D-GNN) combined with Long Short-Term Memory (LSTM) networks to model both the structural dependencies and temporal evolution of microservice behaviors. Unlike traditional approaches, our method uses only CPU utilization as the sole monitoring metric, leveraging its high observability and strong correlation with service performance. From a symmetry perspective, normal microservice behaviors exhibit approximately symmetric spatiotemporal patterns: structurally similar services tend to share similar CPU trajectories, and recurring workload cycles induce quasi-periodic temporal symmetries in utilization signals. Runtime anomalies can therefore be interpreted as symmetry-breaking events that create localized structural and temporal asymmetries in the service graph. The proposed framework is explicitly designed to exploit such symmetry properties: the D-GNN component respects permutation symmetry on the microservice graph while embedding the evolving structural context of each service, and the LSTM module captures shift-invariant temporal trends in CPU usage to highlight asymmetric deviations over time. Experiments conducted on real-world microservice datasets demonstrate that the proposed method delivers excellent performance, achieving 98 percent accuracy and 98 percent F1-score. Compared to baseline methods such as DeepTraLog, which achieves 0.93 precision, 0.978 recall, and 0.954 F1-score, our approach performs competitively, achieving 0.980 precision, 0.980 recall, and 0.980 F1-score. Our results indicate that a single-metric, symmetry-aware spatiotemporal modeling approach can achieve competitive performance without the complexity of multi-metric inputs, providing a lightweight and robust solution for real-time anomaly detection in large-scale microservice environments. Full article

Transfer learning, which utilizes extensive simulation data to overcome the limitations of scarce and expensive experimental data, has emerged as a powerful approach for predictive modeling in various physical domains. This study presents a comprehensive framework to improve the predictive performance of transfer learning, focusing on quasi-zero stiffness (QZS) systems with limited experimental datasets. The proposed framework systematically examines the interplay among three critical factors in the target domain: data augmentation, layer-freezing configurations, and neural network architecture. Simulation-driven synthetic data are generated to capture dynamic features not represented in the sparse experimental data. The optimal transfer depth is explored by evaluating different scenarios of selective layer freezing and fine-tuning. Results show that partial transfer strategies outperform both full-transfer and non-transfer approaches, leading to more stable and accurate predictions. To investigate hierarchical transfer, both symmetric and asymmetric network architectures are designed, embedding physically meaningful representations from simulations into the deeper layers of the target model. Furthermore, an attention mechanism is integrated to emphasize material-specific characteristics. Building on these components, the proposed simulation-driven framework predicts the full force–displacement responses of QZS systems using only 12 experimental samples. Through a systematic comparison of three datasets (direct transfer, linear correction, FEM-based correction), three network architectures, and seven layer-freezing scenarios, the framework achieves a best test performance of R² = 0.978 and MAE = 0.34 Newtons. Full article

We use deep Physics-Informed Neural Networks (PINNs) to simulate stratified forced convection in plane Couette flow. This process is critical for atmospheric boundary layers (ABLs) and oceanic thermoclines under global warming. The buoyancy-augmented energy equation is solved under two boundary conditions: Isolated-Flux (single-wall heating) and Flux–Flux (symmetric dual-wall heating). Stratification is parameterized by the Richardson number $(R_i\in [-1,1\left]\right),$ representing $\pm 2 °\mathrm{C}$ thermal perturbations. We employ a decoupled model (linear velocity profile) valid for low-Re, shear-dominated flow. Consequently, this approach does not capture the full coupled dynamics where buoyancy modifies the velocity field, limiting the results to the laminar regime. Novel contribution: This is the first deep PINN to robustly converge in stiff, buoyancy-coupled flows ($\mid R_i\mid \le 1$) using residual connections, adaptive collocation, and curriculum learning—overcoming standard PINN divergence (errors $>28\%$). The model is validated against analytical ($R_i=0$) and RK4 numerical ($R_i\ne 0$) solutions, achieving $L_2$ errors $\le 0.009\%$ and $L_{\mathrm{\infty }}$ errors $\le 0.023\%$. Results show that stable stratification $(R_i>0)$ suppresses convective transport, significantly reduces local Nusselt number ($Nu$) by up to $100\%$ (driving $Nu$ towards zero at both boundaries), and induces sign reversals and gradient inversions in thermally developing regions. Conversely, destabilizing buoyancy $(R_i<0)$ enhances vertical mixing, resulting in an asymmetric response: $Nu$ increases markedly (by up to $140\%$) at the lower wall but decreases at the upper wall compared to neutral forced convection. At $5–10\times$ lower computational cost than DNS or RK4, this mesh-free PINN framework offers a scalable and energy-efficient tool for subgrid-scale parameterization in general circulation models (GCMs), supporting SDG 13 (Climate Action). Full article

This study focuses on the East Channel Project (Xiang’an South Road—Airport Expressway Section). The project is in the South Port Harbor Bay area. The area has highly complex and asymmetrical geology. Construction faces multiple challenges: tight schedule, overlapping pipeline operations, and large-scale foundation treatment needs. To tackle these, the project uses the plastic drainage board surcharge preloading method for ground improvement. This technique needs continuous settlement deformation monitoring. The monitoring aims to spot potential asymmetric trends and fix the best unloading time. Traditional settlement prediction methods have limits. So, this study develops an intelligent prediction model (SSA-BP). It combines the Sparrow Search Algorithm (SSA) with the BP neural network. The model uses SSA’s strong global search ability to optimize the BP network’s initial weights and thresholds. This effectively avoids local minima and improves prediction stability. Comparative experiments with other optimization algorithms (Particle Swarm Optimization PSO, Grey Wolf Optimizer GWO, and Differential Evolution DE) show that the SSA-BP model has better convergence accuracy and robustness. Field monitoring data validation indicates the model’s prediction error is stably between −3.4% and 3.2%. It surpasses traditional methods like the three-point and hyperbolic methods. The study’s key innovation is introducing an asymmetry-aware view. It analyzes settlement’s morphological evolution and predictability under surcharge preloading. The SSA-BP model can identify both symmetric and asymmetric deformation patterns well. It offers a new computational tool to understand asymmetry breaking in geotechnical systems. Moreover, the model can accurately predict settlement behavior in real time. This provides dynamic construction decision-making guidance and effective cost control. This research shows that intelligent algorithms have great potential. They can reveal complex geotechnical systems’ inherent laws and promote foundation engineering’s intelligentization. Full article

Accurate quality-of-transmission (QoT) estimation prior to lightpath deployment is essential for minimizing design margins in optical networks. Owing to their high precision and strong generalization capabilities, artificial neural networks (ANNs) have emerged as a promising approach for lightpath QoT estimation. However, focusing exclusively on prediction accuracy is inadequate for maximizing global network capacity. Conventional models employing symmetric loss functions apply identical penalties to both overestimation and underestimation errors, thereby precluding controlled bias in predictions and their impact on overall network capacity. This paper investigates the margin configuration for the whole network capacity and proposes a novel QoT estimation method with asymmetric loss functions, which jointly considers the assessment of global network capacity and gives different penalties for overestimation and underestimation. We further present an iterative search algorithm grounded in network capacity considerations to optimize the parameters of these asymmetric loss functions. Simulation results confirm that our ANN-based models facilitate efficient modulation format assignment, leading to corresponding increases in network capacity. Full article