Search Results (27)

A novel modified generalized Riccati equation mapping neural network-based approach is the basic theme of this study by exploring the nonlinear dynamical characteristics of the the strain wave model’s soliton solutions, which govern wave propagation in micro structured solids. Strain waves are particularly intriguing, since they preserve their form and speed throughout transmission. The nonlinear dynamical behaviors of strain waves may be modeled by partial differential equations in micro structured materials. In the realm of micro structured solids, there exists a class of phenomena that are referred to as micro strain waves. These waves arise in solids possessing intricate internal architectures, including periodic lattices, precisely engineered metamaterials Understanding these waves is key to designing more complex materials and new acoustic technologies. The activation function and the weight function of the neural network are assigned to each input layer, hidden layer and output layer and the neural network itself is a multi-layer computational network. Using the structure of the neural network, every neuron in the first hidden layer is given solutions to the Riccati equation, and the new highly expressive trial functions are generated in a systematic way. In this way, a large variety of exact soliton solutions are obtained, such as bright, dark, kink, and combined solitons as well as periodic and hyperbolic wave profiles. The influence of the essential physical and mathematical parameters is explored systematically using three-dimensional, two-dimensional and contour visualizations, which illustrate how parameter variations lead to changes in the amplitude, shape and stability of the wave structures. The solutions presented reveal the dynamic properties of micro strain solitons which leads to new avenues of investigation in the study of related nonlinear phenomena in micro structured solids. In a broader context, our results highlight the great potential of analytical techniques using neural networks as a powerful and versatile toolset to study complex nonlinear wave models within the applied sciences from acoustics to photonics to smart materials engineering. Full article

Graph-based fraud detection in multi-relational social networks must capture heterogeneous relation semantics and diverse fraud patterns while preserving geometric consistency and remaining scalable. Existing methods often either force all relations into a shared Euclidean or single-curvature space, or fuse relation-wise embeddings after mapping them to tangent coordinates, which weakens curvature-dependent metric information. We propose Relation-Specific Curvature Fields on Product Manifolds (RSCF-PM), a geometry-consistent framework that learns relation-specific curvature and represents each node as a tuple on a Riemannian product manifold. Each relation is encoded in its own hyperbolic space, and cross-relation fusion is performed directly through the product metric rather than Euclidean concatenation. On top of this representation, we introduce a multi-prototype classifier to model multiple fraud modes within each class. To support large-scale training, we adopt tangent-space aggregation as an efficient approximation to the Fréchet mean. Experiments on four public fraud detection benchmarks, including the 5.78M-node T-Social network, show that RSCF-PM achieves the best results on T-Social, FDCompCN, and YelpChi, while remaining highly competitive on Amazon, with up to 4.96% AUC improvement over strong baselines. Ablation and efficiency studies further confirm the complementary value of each component and the practical scalability of the framework. Full article

The design of broadband, high-efficiency solar absorbers remains challenging due to the complex and ill-posed inverse mapping from the target optical responses to the physical structures in inverse design optimization. To address this, we propose a joint forward–inverse deep learning framework that enables the rapid and accurate optimization of multilayer metamaterial absorbers. This method integrates an inverse network based on a Modified Swin Transformer with a Multilayer Perceptron forward proxy and performs end-to-end training in a consistency-driven cycle. This strategy reduces the one-to-many ambiguity in inverse design and improves the prediction accuracy, with normalized test mean squared errors of 7.2 × 10⁻⁵ (inverse) and 6.8 × 10⁻⁵ (forward). Using this framework, we optimized an absorber comprising W/SiO₂ hyperbolic metamaterial stacks and TiO₂/SiO₂ anti-reflection coatings, achieving 97.4% average absorptivity across the 400–1750 nm solar spectrum, along with polarization insensitivity and robust wide-angle performance up to 60° incidence. The outdoor solar heating tests showed that the fabricated absorber reaches a peak temperature of 86.3 °C under natural sunlight, with an irradiance peak of about 850 W/m² at noon. This work shows that combining forward and reverse deep learning provides a powerful and scalable paradigm for accelerating the intelligent design of high-performance solar thermal metamaterials. Full article

Graph attention improves neighbor discrimination, but it remains limited by local receptive fields and by a strong dependence on the input topology, which is often unreliable on heterophilous graphs. We propose Geometric Graph Learning Network (G2LNet), a structure-learning framework that infers message-passing probabilities from an explicit geometric topology learned in latent Euclidean or hyperbolic spaces. G2LNet combines (i) a geometric mapping module, (ii) distance- or inner-product-based relation operators with perceptual connectivity to control the influence of the given graph, and (iii) end-to-end constraint objectives enforcing stability, sparsity, and (optional) symmetry of the learned topology. This design yields unified local, non-local, and graph-free neighborhoods, enabling systematic analysis of when non-local aggregation helps. Experiments on node classification across nine publicly available benchmark datasets demonstrate that G2LNet’s controlled variant consistently achieves higher accuracy than representative strong baseline models–both local and non-local–on most datasets. This establishes a robust alternative for smaller scale node classification tasks. Full article

Controlled-release systems must translate material design choices into predictable pharmacokinetic (PK) profiles, yet purely mechanistic or purely data-driven models often underperform when tuning complex polymer networks. Here, we develop tunable poly(vinyl formal) membranes (PVFMs) for diclofenac delivery and integrate classical kinetic analysis with a Generalized Regression Neural Network (GRNN) to connect formulation variables to release behavior and PK-relevant targets. PVFMs were synthesized across a gradient of crosslink densities by varying HCl content; diclofenac release was quantified under standardized conditions with geometry and dosing rigorously controlled (thickness, effective area, surface-area-to-volume ratio, and areal drug loading are reported to ensure reproducibility). Release profiles were fitted to Korsmeyer–Peppas, zero-order, first-order, Higuchi, and hyperbolic tangent models, while a GRNN was trained on material descriptors and time to predict cumulative release and flux, including out-of-sample conditions. Increasing crosslink density monotonically reduced swelling, areal release rate, and overall release efficiency (strong linear trends; r ≈ 0.99) and shifted transport from anomalous to Super Case II at the highest crosslinking. Classical models captured regime transitions but did not sustain high accuracy across the full design space; in contrast, the GRNN delivered superior predictive performance and generalized to conditions absent from training, enabling accurate interpolation/extrapolation of release trajectories. Beyond prior work, we provide a material-to-PK design map in which crosslinking, porosity/tortuosity, and hydrophobicity act as explicit “knobs” to shape burst, flux, and near-zero-order behavior, and we introduce a hybrid framework where mechanistic models guide interpretation while GRNN supplies robust, data-driven prediction for formulation selection. This integrated PVFM–GRNN approach supports rational design and quality control of controlled-release devices for diclofenac and is extendable to other therapeutics given appropriate descriptors and training data. Full article

This paper presents DeepMask-GPR, a novel deep learning framework for automatic detection and geometric estimation of buried cylindrical objects in ground-penetrating radar (GPR) B-scan images. Built upon Mask R-CNN, the proposed method integrates hyperbola detection, apex localization, and real-world coordinate mapping in an end-to-end architecture. A curvature-enhanced dual-channel input improves the visibility of weak hyperbolic patterns, while a quadratic regression loss guides the network to recover precise geometric parameters. DeepMask-GPR eliminates the need for raw signal data or manual post-processing, enabling robust and scalable deployment in field scenarios. On two public datasets, DeepMask-GPR achieves consistently higher TPR/IoU for spatial localization than baselines. On an in-house B-scan set, it attains low MAE/RMSE for radius estimation. Full article

Recent studies have demonstrated the potential of hyperbolic paraboloid (hypar), a doubly curved geometry, in coastal engineering applications. Predicting pressure distribution, critical for subsequent finite element analysis, on such novel three-dimensional structures require Computational Fluid Dynamics (CFD) simulations, which are computationally intensive. To address this challenge, the current study develops an artificial neural network (ANN) surrogate to predict pressure distributions on hypar free-surface breakwaters (FSBWs) under solitary wave loading. Using Smoothed Particle Hydrodynamics (SPH) as the CFD tool, simulations generate the supervised learning dataset, where inputs are the hypar warping $R_n$, breakwater draft $d_r$, and wave height $H$. The targets consist of two $30\times 30$ pressure maps at wave arrival (hydrostatic) and peak, together with the wave rise time $\left\{\mathit{P}\right(t_0)$, $\mathit{P}\left(t_{peak}\right)$, $\mathsf{\Delta }t=t_{peak}-t_0\}$. Three architectures, FNN, CNN, and DeepONet, are trained with homoscedastic uncertainty loss weighting, each at two parameter sizes ($~50\mathrm{k}$ and $~500\mathrm{k}$). Results for training and testing show that all models achieve low errors, with models with ~50k parameters found to be sufficient, and scaling to ~500k yields some generalization improvement. Further reducing the parameters (~5k) degrades accuracy for all models, with DeepONet proven most robust to parameter size reduction. Overall, this study introduces a novel SPH-ANN workflow for predicting wave pressures on hypar FSBWs, where inference on new samples occurs in a few milliseconds per sample, delivering orders-of-magnitude speedups relative to running new SPH simulations. This computational efficiency enables rapid design iteration and optimization of hypar FSBWs, facilitating their potential deployment in coastal defense. Full article

Hysteresis primarily affects the positioning accuracy of the magnetic shape memory alloy-based actuator (M-SMAA). This paper proposes the use of the nonlinear autoregressive moving average with an exogenous input (NARMAX) model to describe the complex dynamic hysteresis of M-SMAA. First, an improved Prandtl–Ishlinskii operator is proposed as the exogenous variable function for the NARMAX model, using a hyperbolic tangent function as the input to the exogenous variable function, to better capture and represent the multivalued mapping hysteresis in M-SMAA. Then, a long short-term memory neural network is introduced to construct the NARMAX model, further optimizing its performance. Finally, the experimental results verify the effectiveness of the model. Full article

While recent advances have successfully integrated neural networks with physical models to derive numerical solutions, there remains a compelling need to obtain exact analytical solutions. The ability to extract closed-form expressions from these models would provide deeper theoretical insights and enhanced predictive capabilities, complementing existing computational techniques. In this paper, we study the nonlinear Gardner equation and the (2+1)-dimensional Zabolotskaya–Khokhlov model, both of which are fundamental nonlinear wave equations with broad applications in various physical contexts. The proposed models have applications in fluid dynamics, describing shallow water waves, internal waves in stratified fluids, and the propagation of nonlinear acoustic beams. This study integrates a modified generalized Riccati equation mapping approach and a novel generalized $\left({\displaystyle \frac{G^{\prime }}{G}}\right)$-expansion method with neural networks for obtaining exact solutions for the suggested nonlinear models. Researchers are currently investigating potential applications of these neural networks to enhance our understanding of complex physical processes and to develop new analytical techniques. The proposed strategies incorporate the solutions of the Riccati problem into neural networks. Neural networks are multi-layer computing approaches including activation and weight functions among neurons in input, hidden, and output layers. Here, the solutions of the Riccati equation are allocated to each neuron in the first hidden layer; thus, new trial functions are established. We evaluate the suggested models, which lead to the construction of exact solutions in different forms, such as kink, dark, bright, singular, and combined solitons, as well as hyperbolic and periodic solutions, in order to verify the mathematical framework of the applied methods. The dynamic properties of certain wave-related solutions have been shown using various three-dimensional, two-dimensional, and contour visualizations. This paper introduces a novel framework for addressing nonlinear partial differential equations, with significant potential applications in various scientific and engineering domains. Full article

The continuous reduction in critical dimensions and the escalating demands for higher throughput are driving motion platforms to operate under increasingly complex conditions, including multi-axis coupling, structural nonlinearities, and time-varying operational scenarios. These complexities make the trade-offs among precision, speed, and robustness increasingly challenging. Traditional Proportional–Integral–Derivative (PID) controllers, which rely on empirical tuning methods, suffer from prolonged trial-and-error cycles and limited transferability, and consequently struggle to maintain optimal performance under these complex working conditions. This paper proposes an adaptive β–Proximal Policy Optimization with Random Network Distillation (β-PPO-RND) parameter optimization within the Prescribed Performance Control (PPC) framework. The adaptive coefficient β is updated based on the temporal change in reward difference, which is clipped and smoothly mapped to a preset range using a hyperbolic tangent function. This mechanism dynamically balances intrinsic and extrinsic rewards—encouraging broader exploration in the early stage and emphasizing performance optimization in the later stage. Experimental validation on a Permanent Magnet Linear Synchronous Motor (PMLSM) platform confirms the effectiveness of the proposed approach. It eliminates the need for manual tuning and enables real-time controller parameter adjustment within the PPC framework, achieving high-precision trajectory tracking and a significant reduction in steady-state error. Experimental results show that the proposed method achieves MAE = 0.135 and RMSE = 0.154, representing approximately 70% reductions compared to the conventional PID controller. Full article

Tea bud targets are typically located in complex environments characterized by multi-scale variations, high density, and strong color resemblance to the background, which pose significant challenges for rapid and accurate detection. To address these issues, this study presents YOLO-LMTB, a lightweight multi-scale detection model based on the YOLOv11n architecture. First, a Multi-scale Edge-Refinement Context Aggregator (MERCA) module is proposed to replace the original C3k2 block in the backbone. MERCA captures multi-scale contextual features through hierarchical receptive field collaboration and refines edge details, thereby significantly improving the perception of fine structures in tea buds. Furthermore, a Dynamic Hyperbolic Token Statistics Transformer (DHTST) module is developed to replace the original PSA block. This module dynamically adjusts feature responses and statistical measures through attention weighting using learnable threshold parameters, effectively enhancing discriminative features while suppressing background interference. Additionally, a Bidirectional Feature Pyramid Network (BiFPN) is introduced to replace the original network structure, enabling the adaptive fusion of semantically rich and spatially precise features via bidirectional cross-scale connections while reducing computational complexity. In the self-built tea bud dataset, experimental results demonstrate that compared to the original model, the YO-LO-LMTB model achieves a 2.9% improvement in precision (P), along with increases of 1.6% and 2.0% in mAP50 and mAP50-95, respectively. Simultaneously, the number of parameters decreased by 28.3%, and the model size reduced by 22.6%. To further validate the effectiveness of the improvement scheme, experiments were also conducted using public datasets. The results demonstrate that each enhancement module can boost the model’s detection performance and exhibits strong generalization capabilities. The model not only excels in multi-scale tea bud detection but also offers a valuable reference for reducing computational complexity, thereby providing a technical foundation for the practical application of intelligent tea-picking systems. Full article

This paper proposes an optimal-sliding-mode-based adaptive dynamic programming (ADP) master–slave synchronous control strategy for the actuator saturation and performance constraints that AUVs face in dynamic recovery. First, by introducing the sliding-mode function into the value function to optimize the state error and its derivative simultaneously, the convergence speed is significantly improved. Second, by designing the performance constraint function to directly map the sliding-mode function, the evolution trajectory of the sliding-mode function is constrained, ensuring the steady-state and transient characteristics. In addition, the hyperbolic tangent function (tanh) is introduced into the value function to project the control inputs into an unconstrained policy domain, thereby eliminating the phase lag inherent in conventional saturation compensation schemes. Finally, the requirement for initial stability is relaxed by constructing a single-critic network to approximate the optimal control policy. The simulation results show that the proposed method has significant advantages in terms of the position and attitude synchronization error convergence rate, steady-state accuracy, and control signal continuity compared with the conventional ADP method. Full article

Ground penetrating radar (GPR) is widely used for subsurface object detection, but manual interpretation of hyperbolic features in B-scan images remains inefficient and error-prone. In addition, traditional forward modeling methods suffer from low computational efficiency and strong dependence on field measurements. To address these challenges, we propose an unsupervised data augmentation framework that utilizes CycleGAN-based model to generate diverse synthetic B-scan images by simulating varying geological parameters and scanning configurations. This approach achieves GPR data forward modeling and enhances the scenario coverage of training data. We then apply the EfficientDet architecture, which incorporates a bidirectional feature pyramid network (BiFPN) for multi-scale feature fusion, to enhance the detection capability of hyperbolic signatures in B-scan images under challenging conditions such as partial occlusions and background noise. The proposed method achieves a mean average precision (mAP) of 0.579 on synthetic datasets, outperforming YOLOv3 and RetinaNet by 16.0% and 23.5%, respectively, while maintaining robust multi-object detection in complex field conditions. Full article

Light has been intensively investigated as a computing medium due to its high-speed propagation and large operation bandwidth. Since the invention of the first laser in 1960, the development of optical computing technologies has presented both challenges and opportunities. Recent advances in artificial intelligence over the past decade have opened up new horizons for optical computing applications. This study presents an end-to-end truth table direct mapping approach using on-chip deep diffractive neural network (D²NN) technology to achieve highly parallel optical logic operations. To enable precise logical operations, we propose an on-chip nonlinear solution leveraging the similarity between the hyperbolic tangent (tanh) function and reverse saturable absorption characteristics of quantum dots. We design and demonstrate a 4-bit on-chip D²NN full adder circuit. The simulation results show that the proposed architecture achieves 100% accuracy for 4-bit full adders across the entire dataset. Full article

The spread of biometric systems utilizing fingerprints has increased the need for advanced spoof detection techniques, but training convolutional neural networks (CNNs) with the limited number of images available in fingerprint datasets poses significant challenges. In this paper, we propose a lightweight network architecture which addresses the challenges inherent in small fingerprint datasets by employing a moderately deep network architecture which is sufficient for extracting essential features from fingerprint images. We apply a hyperbolic tangent activation to the final feature map, which has features from local receptive fields, and average the responses into a single value. Thus, our architecture reduces overfitting by increasing the number of effective labels during training. Additionally, the incorporation of the spatial attention module enhances feature representation, culminating in improved accuracy. The evaluation results show that the proposed model, with only 0.14 million parameters, outperforms existing techniques including lightweight models and transfer-learning-based models, achieving superior average test accuracies of 98.30% and 95.57% on the LivDet-2015 and -2017 datasets, respectively. It also delivers state-of-the-art cross-material performance, with corresponding average classification error values of 0.81% and 1.91%, making it highly effective for on-device fingerprint authentication. Full article