Search Results (100)

Corporate balance sheets report aggregate equity and liability totals but conceal the internal allocation of financing sources across asset categories—an identification problem that conventional econometric methods cannot resolve without additional parametric assumptions. This paper develops a maximum entropy (ME) panel estimator to recover two latent scalar parameters: x ∈ (0,1), the share of equity capital directed toward long-term asset financing, and y ∈ (0,1), the corresponding debt allocation share. Grounded in maximum entropy principle, the estimator selects the unique parameter vector that satisfies the mean-level balance-sheet constraint while maximising joint Shannon entropy—the least-biassed solution consistent with observable data. The closed-form logistic representation yields a scalar Lagrange multiplier λ*, interpreted as a financing pressure index, recoverable via bisection in at most 21 iterations at tolerance ε = 10⁻⁵. Building on the ME estimates, we introduce a continuous matching alignment index M* = x* − y* that measures the degree of compliance with the financial matching principle along a continuous spectrum rather than as a binary categorisation. Applied to a ten-firm, cross-sectoral panel spanning Technology, Finance, Energy, and Automotive sectors over an observation window spanning 2001 to 2025 (with firm-specific subperiods reflecting differences in IPO dates and data availability), the framework reveals substantial heterogeneity in latent financing flows: equity allocation shares range from 30.1% (NVIDIA) to 75.1% (ExxonMobil), while debt allocation shares span 37.1% to 77.5%. Across the panel, only Meta exhibits substantial positive matching alignment, while Microsoft, ExxonMobil, Apple, and Tesla show only very slight differences that fall within the neutral band, and the remaining firms show varying degrees of structural departure from the matching benchmark; the thresholds used to summarise these descriptive labels are interpretive aids rather than re-imposed binary criteria, and the substantive ranking of firms along M* does not depend on the specific threshold values adopted. The ME solution’s entropy H(x*, y*) and the normalised diversification index D(x*, y*) describe allocation balance under the estimator’s information–theoretic criterion rather than independently observed firm complexity; in the present sample, the cross-firm ordering of these values is not recovered by firm size, leverage, or sector classification alone. These findings, based on a ten-firm case-study panel with time-invariant allocation parameters, should be interpreted as descriptive patterns of the present sample rather than statistically validated regularities. They provide a theoretically rigorous and computationally tractable identification of unobservable corporate financing flows, with potential implications for capital structure theory, financial risk assessment, and balance sheet analysis that would benefit from validation on larger and more representative samples in future work. Full article

Intelligent sensors and Internet of Medical Things (IoMT) platforms are rapidly changing smart healthcare by enabling continuous capture of physiological, behavioral, and clinical events outside conventional hospital settings. Yet the value of connected sensing depends on more than signal acquisition alone. A practical remote-monitoring ecosystem must also convert sensor alerts, clinician-facing summaries, and historical electronic clinical records (ECRs) into ranked evidence that supports care decisions. This study reframes a large-AI clinical retrieval model as the intelligence layer of an edge–cloud IoMT architecture. The proposed framework combines Transformer-Based Sequence (TBS) encoding, BioBERT-driven representation learning, explicit retrieval, and domain-guided re-ranking to connect sensor-originated narratives, patient records, and clinician queries. The empirical evaluation is conducted on Medical Information Mart for Intensive Care III (MIMIC-III) and i2b2, two de-identified clinical text benchmarks that approximate the documentation layer of real-world remote patient monitoring. Compared with strong baselines, including DeepBio, UniT2T, Web4IR, A2A-API, CoLTiD, VLRG, ColBERT, DeepSDH, BiRex, and DL4BTM, the proposed model achieves the best overall performance, reaching F1/Pre/NDCG scores of 0.8399/0.8338/0.5235 on MIMIC-III and 0.8090/0.8100/0.5129 on i2b2. Ablation experiments confirm the importance of exploratory data adaptation, critical feature modeling, critical token learning, cross-disciplinary supervision, and data-driven regularization. Parameter sensitivity analysis shows stable behavior for beta values greater than or equal to 1, with the strongest results at beta = 5. The study concludes that large-AI retrieval can strengthen the clinical interpretation layer required for IoMT-enabled remote monitoring, while future work should validate the approach on live multimodal sensor streams and privacy-preserving deployments. Full article

In industrial inspection, subtle defects often appear as local variations in appearance or geometry, making reliable anomaly detection challenging. A single sensing modality can miss important defect cues, while multimodal inspection combines appearance and geometric information to represent industrial objects more comprehensively. Many existing multimodal anomaly detection methods adopt early fusion strategies that integrate features at an early stage of the network. Such early integration can dilute modality-specific anomaly responses and cause anomaly smoothing, leading to degraded detection and localization performance. To address these challenges, we propose a reconstruction-based unsupervised multimodal anomaly detection framework integrating Discrepancy-Guided Complementary Fusion (DGCF) and Noise to Feature (N2F). Specifically, DGCF reduces anomaly smoothing by exploiting cross-modal discrepancies to extract complementary information, rather than directly summing or concatenating features from different modalities. Furthermore, N2F injects Gaussian noise into the feature space to regularize feature reconstruction and encourage the decoder to learn robust normal representations. Experimental results on the MVTec 3D-AD and Eyecandies datasets demonstrate the effectiveness of the proposed method. The proposed method achieves 97.3% I-AUROC, 99.6% P-AUROC, and 97.6% AUPRO on MVTec 3D-AD, and 94.8% I-AUROC, 98.6% P-AUROC, and 93.4% AUPRO on Eyecandies. Full article

Accurate tree counting from remote sensing data is essential for forest inventory, biomass estimation, carbon accounting, and ecological monitoring. However, existing approaches predominantly rely on airborne RGB imagery and often struggle in complex forest scenes where neighboring crowns exhibit highly similar textures and colors and where overlapping crown boundaries become ambiguous. To address this limitation, the LiDAR-derived Canopy Height Model (CHM) is introduced as a complementary modality that provides explicit cues on canopy height variation and vertical structure to support RGB-based analysis. Building on this, we propose BCAR-Net, a broker-guided RGB and depth (RGB-D) multimodal framework that couples bidirectional cross-modal interaction, adaptive tri-branch fusion, and auxiliary reconstruction within a two-stage optimization scheme. Specifically, a bidirectional cross-attention U-Net generates an intermediate broker RGB-D representation from paired RGB images and depth maps through symmetric bidirectional cross-attention between the two modalities and direction-aware gating. The original RGB image, depth map, and broker representation are then jointly encoded by three weight-sharing branches and adaptively aggregated by a spatial fusion gate for density-map regression. To regularize the fused latent feature, a multi-scale cross-attention reconstruction decoder provides auxiliary RGB and depth reconstruction supervision by querying multi-scale BCA-UNet encoder features through 2D cross-attention, and a reconstruction-oriented first stage replaces externally generated fused-image supervision, yielding a task-consistent optimization scheme. Experiments on the NEONTreeEvaluation benchmark show that BCAR-Net consistently outperforms single-modality settings and direct RGB-D concatenation multimodal baseline. Additional experiments on a public UAV RGB–LiDAR dataset provide a small-scale supplementary evaluation under a different acquisition setting, where BCAR-Net achieves modest but consistent improvements over RGB-only and depth-only baselines. These results demonstrate that the proposed framework offers an effective but computationally cautious solution for tree counting in complex forest environments. Full article

A coupled multi-stage learning framework is presented for joint bit error rate (BER) prediction and beam angle optimization in massive multiple-input multiple-output (MIMO) systems under a controlled simulation protocol. Unlike purely sequential benchmarking pipelines, the proposed method jointly coordinates BER prediction and beam-angle selection through a shared latent representation, an uncertainty-guided refinement mechanism, a cross-stage consistency loss and alternating optimization. Ten diverse approaches are systematically evaluated across two task-specific stages: Stage 1 examines six classical and adapted methods for BER prediction, including polynomial regression and deep unfolding networks; Stage 2 investigates four machine-learning and generative adversarial network (GAN)-based approaches for angle optimization, including conditional GANs and the proposed Direct-Angle neural network. Stage 3 couples the best-performing methods into a unified hybrid architecture through a shared encoder, explicit consistency regularization and alternating cross-stage updates, thereby producing an integrated beamforming decision strategy rather than an independent cascade. It is shown through the evaluation that the coupled hybrid framework achieves 96.0% overall angle-selection accuracy, a mean BER of $8.0\times {10}^{-5}$ and 100% BER tolerance compliance within $\pm 3$ dB. In this framework, a differentiable BER surrogate initialized from a second-degree polynomial-regression teacher is coupled with the proposed Direct-Angle-NN for angle optimization. Relative to the strongest reimplemented literature baseline under the same controlled simulation assumptions, a 33.3% reduction in mean BER is achieved. Ablation experiments show that the coupling mechanism provides a modest but consistent improvement over the decoupled sequential baseline, increasing angle-selection accuracy from 93.5% to 96.0% and reducing mean BER from $1.05\times {10}^{-4}$ to $8.0\times {10}^{-5}$; the shared encoder accounts for the largest part of this gain while the consistency loss adds 0.6 percentage points. These results indicate that the shared encoder, consistency regularization and uncertainty-guided refinement improve the final beamforming decision, although the gain should be interpreted as incremental rather than as a large architectural breakthrough. A spectral efficiency of 38.0 bps/Hz and an energy efficiency of 0.466 Gbps/W are achieved with a power consumption of only 32.6 W. The theoretical discussion is presented as an analytical characterization of BER sensitivity, complemented by a computational-complexity assessment and empirical convergence diagnostics for the alternating optimization, rather than as a formal optimality proof. The effectiveness of the framework across multiple performance metrics is supported by Monte Carlo simulations, while the limitations of the current setup, including perfect CSI, uncoded QPSK, ideal hardware assumptions and a fixed beam codebook, are explicitly discussed. The complete simulation framework, including code and trained models, can be made available by the corresponding author upon reasonable request to facilitate reproducible research in massive MIMO optimization. Full article

Securing IoT networks presents fundamental challenges rooted in hardware constraints: firmware is often non-upgradeable and every security boundary is fixed at manufacture. Machine learning-based intrusion detection offers a scalable response, yet nearly all published systems assume clean training data and clean inference conditions. Production IoT environments satisfy neither assumption. Sensors degrade, packets drop, and adversaries deliberately corrupt telemetry streams to evade detection. The framework described here is built around that reality. The proposed framework is distinguished from prior work by four design decisions. First, three encoding branches, a residual DNN, a 1D-CNN, and a BiLSTM, are run in parallel and are fused by concatenation, each capturing structural patterns in tabular traffic data that the others miss. Second, a dual-view consistency loss trains the model under simultaneous feature masking and Gaussian noise, penalizing prediction divergence between two independently corrupted views of the same sample. Third, we introduce entropy-weighted attention: rather than fixed learned weights, per-feature importance is adjusted dynamically from information entropy measured across training batches, giving higher-entropy features stronger influence because they carry more discriminative variation. Fourth, branch-dropout regularization randomly silences entire branches during training, forcing each to develop independently useful representations instead of co-adapting. Class imbalance is handled through severity-aware loss weighting which scales contributions by the operational cost of missing each attack category, not purely by inverse frequency. On UNSW-NB15, the full model achieves 99.99% accuracy, 100% precision, 99.97% recall, and a false-negative rate of 2.65 × 10⁻⁴—the lowest across all compared architectures. Full article

Alzheimer’s disease (AD) is characterized by pronounced spatial heterogeneity and complex neurodegenerative patterns, which pose significant challenges for representation learning on three-dimensional brain images. Conventional convolutional neural networks relying on regular grid sampling struggle to align localized structural degeneration, and their performance is further compromised by class imbalance and a high proportion of weakly discriminative samples, leading to suboptimal optimization dynamics and reduced generalization ability. To address the aforementioned challenges, this study proposes Deformable Attention and Risk-aware 3D Network (DAR-3DNet), a modeling framework for 3D magnetic resonance imaging (MRI) classification. The proposed method incorporates deformable sampling and sampling-point modulation into spatial attention generation, enabling the attention estimation process to better adapt to the non-rigid spatial patterns associated with brain structural degeneration. On this basis, an instance-adaptive label smoothing loss with composite risk, termed Instance-wise Adaptive Label Smoothing Loss with Composite Risk (IASLCR), is further introduced to dynamically adjust supervision strength based on sample-specific risk, thereby alleviating optimization bias caused by class imbalance and weakly discriminative samples. Experiments conducted on 1749 structural MRI scans from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset show that the proposed method achieves accuracy values above 0.93 on the AD vs. normal control (NC), mild cognitive impairment (MCI) vs. NC, and AD vs. MCI classification tasks, while yielding better overall performance than the evaluated baseline models. These results suggest that the proposed framework has considerable potential for structural MRI-based AD/MCI classification. Full article

While Graph Autoencoders (GAEs) have become a standard for unsupervised representation learning, their reliance on integer-order convolutions inherently restricts information propagation to immediate local neighborhoods. This paper introduces the Fractional Graph Autoencoder (FGAE) and its variational extension (FVGAE) to move beyond these local constraints. By integrating fractional Laplace operators, our framework generalizes conventional GAEs and enables tunable non-local propagation. We show that the fractional order $\alpha$ acts as a structural regularizer, utilizing the Green’s function of anomalous diffusion to induce a form of structural memory within the latent space. This allows the model to recover long-range dependencies that are typically lost in standard architectures. Systematic benchmarking across eight datasets—ranging from homophilic citation networks to heterophilic and dense product graphs—shows that these fractional variants consistently outperform both foundational and state-of-the-art baselines (ARGA, SIG-VAE, and GraphMAE). Notably, on the Amazon Computers and Citeseer datasets, our methods achieve relative increases in Normalized Mutual Information (NMI) of 77.55% and 67.28%, respectively. Statistical analysis confirms these gains are robust, with large effect sizes (Cohen’s $d>0.80$) and significance at $p<0.05$. These findings suggest that fractional graph autoencoding offers a mathematically grounded inductive bias for capturing the complex, multi-scale dynamics of real-world networked systems. Full article

Synthetic Aperture Radar (SAR) images are affected by speckle noise, which limits their application in fine object interpretation and quantitative analysis. Recent deep learning-based single-image SAR despeckling methods have made significant progress in spatial structure modeling but struggle to exploit temporal redundancy in multi-temporal data. Existing multi-temporal despeckling methods usually rely on complex spatiotemporal network structures, which are prone to overfitting or excessive smoothing of details when training samples are limited. To address these challenges, this paper proposes a latent-space-guided multi-temporal SAR despeckling method from the perspective of transfer learning and representation alignment, achieving effective knowledge transfer from single-image SAR despeckling to multi-temporal despeckling tasks. The method treats the single-image SAR despeckling task as a knowledge source domain, using stable latent space representations learned from the pre-trained single-image despeckling model as prior constraints. A latent space regularization mechanism is introduced during the training of the multi-temporal despeckling model, thereby establishing an explicit representation bridge between the 2D spatial model and the 3D spatiotemporal model. With this strategy, the multi-temporal model inherits the structural perception capability of the single-image model under limited training samples, improving speckle suppression while effectively maintaining image detail and structural consistency. Additionally, a pure convolutional network architecture is employed to support variable-length multi-temporal sequence input, enhancing the method’s adaptability under different temporal sampling conditions. Full article

We study fractional and complex powers of a fixed directional derivative in ${\mathbb{R}}^d$, defined via a Marchaud-type singular integral representation. Under explicit convergence assumptions, this yields a pointwise nonlocal realization along rays. We then formulate a Ramanujan–Hardy approach to fractional directional differentiation based on analytic interpolation of the directional jet at a point. This construction is local in jet space and is governed by Hardy’s formulation of Ramanujan’s Master Theorem. We emphasize that the resulting Ramanujan–Hardy derivative is defined through a Hardy-admissible interpolant of the directional jet. As an application, we investigate fractional directional derivatives of the Newtonian kernel in dimension $d\ge 3$. After a justified regularization and reduction to a Marchaud-type integral, we obtain a one-dimensional integral representation and a zonal harmonic description of the resulting function. This leads to a fractional Maxwell–Gegenbauer identity for $0<\Re \left(s\right)<1$, expressing the fractional directional derivative of ${\parallel x\parallel }^{2-d}$ in terms of Gegenbauer functions of complex degree. In this way, the classical Maxwell multipole formula appears as the integer-order case of a continuous analytic family. Moreover, the fractional operator preserves the main structural properties of the Newtonian kernel, including homogeneity, rotational invariance, and harmonicity away from the origin. The paper thus connects Mellin analysis, Ramanujan’s Master Theorem, fractional calculus, and harmonic analysis on the sphere, while clarifying the distinction between Marchaud and jet-interpolation constructions of fractional directional operators. Full article

Let $U,V\subset \mathbb{H}$ be bounded $C^1$ domains, and let f be quaternion-valued on $U\times V$. We study the mixed Cauchy–Fueter system $D_x^{L}f=0$ and $f{D}_y^R=0$ on product domains by iterating the classical one-variable Borel–Pompeiu formulas in an order consistent with quaternionic multiplication. Under closure regularity on $\overline{U}\times \overline{V}$, we prove an iterated representation formula and show that, in the biregular case, the boundary contribution reduces to the distinguished boundary $\partial U\times \partial V$. This leads to a distinguished boundary transform, ${\mathcal{T}}_{U,V}$, on continuous boundary data. We prove that ${\mathcal{T}}_{U,V}$ maps $C(\partial U\times \partial V;\mathbb{H})$ into $C^{\infty }(U\times V;\mathbb{H})$, establish compact subset estimates for mixed real derivatives, and derive a local approximation theorem within the transform range by finite sums of separated one-variable Cauchy transforms. The analysis is restricted to this representation framework. In particular, the paper does not address a general solvability theory for the mixed inhomogeneous system and does not characterize the full range of ${\mathcal{T}}_{U,V}$. Full article

Chromosome foreground segmentation is a binary semantic segmentation problem that serves as a prerequisite for overlap reasoning, contact-region inspection, and automated karyotyping. Although simpler than full instance separation in formulation, it remains difficult in metaphase imagery because chromosomes are elongated, deformable, weakly bounded, and frequently touching or partially overlapping. To address these chromosome-specific difficulties, we present DCN-KUnet as a task-oriented integration rather than a new generic segmentation family. The encoder–decoder backbone embeds DCNv3 modules to perform geometry-adaptive sampling for bending-aware and boundary-aware representation learning, while a B-spline KAN bottleneck refines the compressed semantic representation through lightweight nonlinear transformation. In addition, a hybrid objective composed of mask supervision, semantic consistency regularization, and internal feature regularization (${\mathcal{L}}_{cd}+{\mathcal{L}}_{SCR}+{\mathcal{L}}_{IFD}$) jointly constrains prediction accuracy, cross-stage semantic agreement, and feature compactness during training. Experiments on the public overlapping-chromosome dataset and on AutoKary2022 converted to binary foreground masks show that DCN-KUnet achieves stronger Dice, IoU, and HD95 with a moderate parameter budget. These results support the proposed framework as a practical and lightweight semantic foreground front-end for chromosome analysis pipelines rather than a full instance-disentanglement solution. Full article

Referring Remote Sensing Image Segmentation (RRSIS) is a representative multimodal understanding task for remote sensing, which segments designated targets from remote images according to free-form natural language descriptions. However, complex remote sensing characteristics, such as cluttered backgrounds, large-scale variations, small scattered targets and repetitive textures, lead to unstable visual grounding and further spatial grounding drift, resulting in inaccurate segmentation results. Existing approaches typically perform implicit visual–linguistic fusion across encoding and decoding stages, entangling spatial grounding with mask refinement. This tightly coupled formulation lacks explicit structural constraints and is prone to cross-modal ambiguity, especially in complex remote sensing layouts. To address these limitations, we propose a Structurally consistent and Grounding-aware Stagewise Reasoning Framework (SGSRF) that follows a grounding-first, segmentation-second paradigm. The framework decomposes inference into three cascaded stages with progressively imposed structural constraints. First, Cross-modal Consistency Refinement (CCR) lays the foundation for stable spatial grounding by enhancing visual–textual structural alignment via CLIP-based features and Structural Consistency Regularization (SCR), producing well-aligned multimodal representations and reliable grounding cues. Second, Grounding-aware Prompt (GPG) Generation bridges grounding and segmentation by converting aligned representations into complementary sparse and dense prompts, which serve as explicit grounding guidance for the segmentation model. Third, Grounding Modulated Segmentation (GMS) leverages the Segment Anything Model (SAM) to generate fine-grained mask prediction under the joint guidance of prompts and grounding cues, improving spatial grounding stability and robustness to background interference and scale variation. Extensive experiments on three remote sensing benchmarks, namely RefSegRS, RRSIS-D, and RISBench, demonstrate that SGSRF achieves state-of-the-art performance. The proposed stagewise paradigm integrates structural alignment, explicit grounding, and prompt-driven segmentation into a unified framework, providing a practical and robust solution for RRSIS in real-world Earth observation applications. Full article

As modern power systems grow increasingly complex, there is a pressing need for stability analysis methods capable of handling nonlinear dynamics while providing physically meaningful and reliable stability indices. Port-Hamiltonian (PH) frameworks have emerged as strong candidates in this regard, offering inherently stable formulations, energy-consistent representations, and modular plug-and-play scalability. However, the practical deployment of PH-based stability analysis remains hindered by the absence of reliable, high-fidelity parameter identification methods that rely on sensor measurements to capture system dynamics while remaining compatible with PH model structures. This paper addresses that gap by proposing a comprehensive three-stage data-driven identification framework for PH modeling of synchronous generators—the central dynamic component of any power system. While the IEEE Standard 115 provides established procedures for transient parameter identification, it exhibits fundamental limitations when applied to PH modeling, including single-scenario identifiability constraints, noise-sensitive derivative-based formulations that amplify sensor measurement errors, and the inability to decouple generator-internal damping from grid contributions. The proposed framework resolves these limitations through multi-scenario excitation using sensor-acquired voltage and current signals, derivative-free state consistency optimization, and physics-based regularization that enforces PH structure preservation. Complete identification of eight key parameters (H, D, $X_d$, $X_q$, $X_d^{\prime }$, $X_q^{\prime }$, $T_{do}^{\prime }$, $T_{qo}^{\prime }$) is achieved with errors ranging from 1.26% to 9.10%, and validation confirms RMS rotor angle errors below 1.2° and speed errors below 0.15%, demonstrating suitability for transient stability analysis, passivity-based control design, and oscillation damping assessment. Full article

Radio-frequency (RF) impairments such as thermal noise, phase noise, and nonlinear distortion are inherently coupled in practical multiple-input multiple-output (MIMO) transceivers, yet most existing mitigation techniques treat them independently or rely on correlation-based black-box learning models. These approaches often fail to generalize under varying operating conditions because they do not capture the underlying causal relationships among hardware impairments. This paper proposes a causal representation learning framework that jointly models and mitigates coupled RF impairments by learning disentangled latent variables aligned with their physical causal structure. A causal variational autoencoder with a structured physics-informed prior and causal regularization is developed to recover impairment-specific representations and enable targeted compensation under diverse channel conditions. The framework is evaluated in a controlled MIMO simulation environment to systematically analyze impairment interactions and mitigation performance. Experimental results show that the proposed method significantly outperforms both classical receivers and conventional learning-based approaches. In particular, the framework achieves an average BER reduction of approximately 57% compared with the classical model-based receiver and about 30% relative to correlation-based deep learning models, while also outperforming recent variational autoencoder-based MIMO detectors in robustness under unseen operating conditions. The output signal-to-noise ratio improves by up to 2.2 dB across the evaluated SNR range. Furthermore, latent representation analysis shows a substantial reduction in cross-covariance, with the disentanglement score decreasing from above 0.48 in standard variational models to approximately 0.12 using the proposed causal approach. Under unseen combinations of SNR and impairment severity, the proposed model achieves the lowest BER degradation and a robustness score of 0.86, confirming improved generalization beyond the training distribution. These results demonstrate that causal representation learning provides a principled and effective solution for modeling and mitigating coupled RF impairments in MIMO communication systems. Full article