Search Results (542)

Zero-inflated count data are characterized by an excessive frequency of zeros that cannot be adequately analyzed by a single distribution, such as Poisson or negative binomial. This problem is pervasive in many practical applications, including document–keyword matrix derived from text corpora, where most keyword frequencies are zero. Conventional statistical approaches, such as the zero-inflated Poisson (ZIP) and zero-inflated negative binomial (ZINB) models, explicitly separate a structural zero component from a count component, but they typically assume independent observations and can be unstable when covariates are high-dimensional and sparse. To address these limitations, this paper proposes a graph-based zero-inflated learning framework that combines simple graph convolution (SGC) with zero-inflated count regression heads such as ZIP and ZINB. We first construct an observation graph by connecting similar samples, and then apply SGC to propagate and smooth features over the graph, producing convolutional representations that incorporate neighborhood information while remaining computationally lightweight. The resulting representations are used as covariates in ZIP and ZINB heads, which preserve probabilistic interpretability through maximum likelihood learning. Our experiments on simulated zero-inflated datasets with controlled zero ratios demonstrate that the proposed ZIP+SGC and ZINB+SGC consistently reduce prediction errors compared with their non-graph baselines, as measured by mean absolute error and root mean squared error. Overall, the proposed approach provides an efficient and interpretable way to integrate graph neural computation with zero-inflated modeling for sparse count prediction problems. Full article

To address the limited environmental perception capability and the difficulty of achieving consistent and efficient representation of the workspace feasible domain caused by high dust concentration, uneven illumination, and enclosed spaces in underground coal mines, this paper proposes a digital spatial construction and representation method for underground environments by integrating RGB-D depth vision with ORB-SLAM3. First, a ChArUco calibration board with embedded ArUco markers is adopted to perform high-precision calibration of the RGB-D camera, improving the reliability of geometric parameters under weak-texture and non-uniform lighting conditions. On this basis, a “dense–sparse cooperative” OAK-DenseMapper Pro module is further developed; the module improves point-cloud generation using a mathematical projection model, and combines enhanced stereo matching with multi-stage depth filtering to achieve high-quality dense point-cloud reconstruction from RGB-D observations. The dense point cloud is then converted into a probabilistic octree occupancy map, where voxel-wise incremental updates are performed for observed space while unknown regions are retained, enabling a memory-efficient and scalable 3D feasible-space representation. Experiments are conducted in multiple representative coal-mine tunnel scenarios; compared with the original ORB-SLAM3, the number of points in dense mapping increases by approximately 38% on average; in trajectory evaluation on the TUM dataset, the root mean square error, mean error, and median error of the absolute pose error are reduced by 7.7%, 7.1%, and 10%, respectively; after converting the dense point cloud to an octree, the map memory footprint is only about 0.5% of the original point cloud, with a single conversion time of approximately 0.75 s. The experimental results demonstrate that, while ensuring accuracy, the proposed method achieves real-time, efficient, and consistent representation of the 3D feasible domain in complex underground environments, providing a reliable digital spatial foundation for path planning, safe obstacle avoidance, and autonomous operation. Full article

High-quality commit messages are essential software artifacts because they succinctly communicate the intent and scope of code changes, yet large language models (LLMs) often fail to reflect project-specific writing conventions when used in a zero-shot setting without contextual signals. This study investigates not whether retrieval helps, but how the same retrieved example, when represented differently in the prompt, quantitatively changes generation outcomes. We implement a retrieve-then-generate framework where the target commit’s diff is used as a query for BM25 (Best Matching 25)-based sparse retrieval over a commit-level database, and the top-1 similar commit is optionally injected as an example context. We compare a no-context condition (K = 0) against a minimal-context condition (K = 1) under three context representations: Diff-only, Message-only, and Diff+Message pair. Using Qwen-7B on 8000 evaluation samples with a fixed prompt skeleton, deterministic decoding, and identical post-processing across conditions, we observe negligible differences at K = 0 (BLEU-4 1.14, ROUGE-L 7.47–7.48, METEOR 4.88–4.91), establishing a stable baseline. At K = 1, the same top-1 retrieved case yields systematically different metric responses depending on how it is represented (Diff-only, Message-only, or Diff+Message), even under an identical prompt skeleton, deterministic decoding, and identical post-processing. This indicates that “context representation” is not a cosmetic formatting choice but a first-class prompt-design variable in retrieval-augmented in-context learning for commit message generation. Accordingly, practitioners should select the representation based on the intended objective (e.g., lexical/style alignment vs. change-intent grounding), rather than assuming a universally optimal format. Full article

Remote sensing plays a vital role in monitoring environmental change in Antarctica, offering non-invasive insights into ice dynamics, biodiversity, and fragile ecosystems. Harsh conditions, limited field access, and logistical challenges result in sparse, noisy, and often unlabelled datasets, posing major obstacles for machine learning (ML) approaches. Data scarcity remains a fundamental challenge for uncrewed aerial vehicle (UAV)-based ecological monitoring. While ML models in other Earth observation domains demonstrate state-of-the-art performance, their applicability in Antarctic and polar regions’ settings is limited. This paper reviews the intersection of ML and UAV-based remote sensing in Antarctica under extreme data constraints. We surveyed recent strategies designed to overcome these limitations, including self-supervised learning, physics-informed modelling, and foundation models. Results highlight a notable gap, as polar environments remain excluded from global datasets and benchmarks due to the extensive data requirements of large-scale models. Opportunities exist where multimodal and multi-scale generalisation can enhance cross-domain adaption to data-scarce use cases. Unlike prior reviews on general remote sensing or task-specific polar studies, this work uniquely underscores the need for Antarctic representation in global ML advances, positioning Antarctica as a frontier testbed for machine learning in extreme, inaccessible, and under-resourced fields. Full article

Accurate detection of small objects in remote sensing imagery remains challenging due to their limited texture, sparse features, and weak contrast. To address this, an enhanced small object detection model integrating top–down and bottom–up attention mechanisms is proposed. First, we design two statistical model-constrained feature pre-extraction networks to enhance the spatial patterns of small objects before feeding them into the backbone network. Next, a top–down attention mechanism followed by an overview-then-refinement process is employed to guide region-level feature extraction. Finally, a bottom–up feature fusion strategy is utilized to integrate micro features and macro structural features in a bottom–up manner, enhancing the representational capacity of limited features for small objects. Evaluations on the AI-TOD and SODA-A datasets show that our method outperforms existing benchmark models. On the AI-TOD dataset, it improves AP and AP0.5 by 0.3% and 2.7%, respectively. More notably, on the more challenging SODA-A dataset, it achieves significant gains of 0.5% in AP and 1.4% in AP0.5. These consistent enhancements across different datasets verify the effectiveness of our method in boosting detection performance, particularly for small targets. Full article

Traditional lie detection technologies, such as the polygraph and event-related potential (ERP)-based approaches, often face limitations in real-world applicability due to their sensitivity to psychological states and the complex, nonlinear nature of electroencephalogram (EEG) signals. In this study, we propose SparseMambaNet, a novel neural architecture that integrates the recently developed Bi-Mamba model with a Sparsely Activated Mixture of Experts (MoE) structure to effectively model the intricate spatio-temporal dynamics of EEG data. By leveraging the near-linear computational complexity of Mamba and the bidirectional contextual modeling of Bi-Mamba, the proposed framework efficiently processes long EEG sequences while maximizing representational power through the selective activation of expert networks tailored to diverse input characteristics. Experiments were conducted with 46 healthy subjects using a simulated criminal scenario based on the Comparison Question Technique (CQT) with monetary incentives to induce realistic psychological tension. We extracted nine statistical and neural complexity features, including Hjorth parameters, Sample Entropy, and Spectral Entropy. The results demonstrated that Sample entropy and Hjorth parameters achieved exceptional classification performance, recording F1 scores of 0.9963 and 0.9935, respectively. Statistical analyses further revealed that the post-response “answer” interval provided significantly higher discriminative power compared to the “question” interval. Furthermore, channel-level analysis identified core neural loci for deception in the frontal and fronto-central regions, specifically at channels E54 and E63. These findings suggest that SparseMambaNet offers a highly efficient and precise solution for EEG-based lie detection, providing a robust foundation for the development of personalized brain–computer interface (BCI) systems in forensic and clinical settings. Full article

Accurately forecasting demand for intermittent items is essential for effective inventory control, improved service levels, and cost reduction. This study focuses on highly sparse, irregular, and volatile demand patterns and proposes a generalizable multi-source data-driven framework for intermittent demand forecasting, using automotive spare parts as a representative application scenario. The proposed framework integrates Transformer networks, multi-graph convolutional networks (GCNs), and a Mamba-based feature fusion module. The Transformer captures long-term temporal dependencies in historical demand sequences, while the multi-graph GCN incorporates prior knowledge—including traffic geography, socioeconomic indicators, and environmental attributes—to model spatial correlations across multiple supply nodes. The Mamba-based fusion module then integrates temporal and spatial features into a unified representation, enhancing predictive accuracy and robustness. Extensive experiments on real-world datasets of automotive spare parts in China show that the proposed framework exhibits competitive and often superior performance compared with TiDE, FSNet, Informer, and DLinear across multiple forecasting horizons (3-, 6-, and 9-step), as measured by RMSE, MAE, and $R^2$. The proposed approach provides a practical and adaptable solution for forecasting intermittent demand, offering valuable support for dynamic inventory management. Full article

Background: MicroRNAs (miRNAs) are an important class of non-coding RNAs that regulate gene expression by binding to target mRNAs and influencing cellular processes such as differentiation, proliferation, and apoptosis. Dysregulation in miRNA expression has been reported to be implicated in many human diseases, including cancer, cardiovascular, and neurodegenerative disorders. Identifying disease-related miRNAs is therefore essential for understanding disease mechanisms and supporting biomarker discovery, but time and cost of experimental validation are the main limitations. Methods: We present a graph-based learning framework that models the complex relationships between miRNAs, diseases, and related biological entities within a heterogeneous network. The model employs a message-passing neural architecture to learn structured embeddings from multiple node and edge types, integrating biological priors from curated resources. This network representation enables the inference of novel miRNA–disease associations, even in sparsely annotated regions of the network. The approach was trained and validated on a dataset benchmark using ten replicated experiments to ensure robustness. Results: The method achieved an average AUC–ROC of ~98%, outperforming previously reported computational approaches on the same dataset. Moreover, predictions were consistent across validation folds and robustness analyses were conducted to evaluate stability and highlight the most important information. Conclusions: Integrating heterogeneous biological information and representing it through graph neural network representation learning offers a powerful and generalizable way to predict relevant associations, including miRNA–disease, and provide a robust computational framework to support biomedical discovery and translational research. Full article

In remote sensing and low-altitude unmanned aerial vehicle(UAV) detection scenarios, small target detection is extremely challenging due to the low pixel proportion, sparse features, and complex backgrounds of targets. The reliability of low-altitude security, in particular, is directly dependent on the accuracy of small target detection. However, current methods still face three major limitations: insufficient detection accuracy for targets smaller than 20 pixels; artifacts and false textures introduced by Generative Adversarial Network-based enhancement, which lead to increased false detection rates; and the reliance of existing approaches on specialized architectures, resulting in weak generalization capability and difficulty in adapting to multi-scenario deployment requirements. To address these issues, this paper proposes a plug-and-play dual-mechanism collaborative enhancement framework named HD-BSNet. Firstly, a High-Frequency Differential Perception mechanism is designed to enhance the detailed feature representation of small targets. Secondly, a Background Semantic Modeling mechanism is introduced to learn key features that distinguish targets from the background. Additionally, a Parallel Multi-Scale Focus Module is constructed to further reinforce target features. Extensive experiments on three small target datasets demonstrate that the proposed method effectively improves the accuracy and generalization ability of small target detection. Full article

Intrinsic motivation provides a principled mechanism for driving exploration in reinforcement learning when external rewards are sparse or delayed. A central challenge, however, lies in defining meaningful novelty signals in high-dimensional and combinatorial state spaces, where observation-level density estimation and prediction-error heuristics often become unreliable. In this work, we propose an information-theoretic framework for intrinsically motivated reinforcement learning grounded in the Information Bottleneck principle. Our approach learns compact latent state representations by explicitly balancing the compression of observations and the preservation of predictive information about future state transitions. Within this bottlenecked latent space, intrinsic rewards are defined through information-theoretic quantities that characterize the novelty of state–action transitions in terms of mutual information, rather than raw observation dissimilarity. To enable scalable estimation in continuous and high-dimensional settings, we employ neural mutual information estimators that avoid explicit density modeling and contrastive objectives based on the construction of positive–negative pairs. We evaluate the proposed method on two representative combinatorial routing problems, the Travelling Salesman Problem and the Split Delivery Vehicle Routing Problem, formulated as Markov decision processes with sparse terminal rewards. These problems serve as controlled testbeds for studying exploration and representation learning under long-horizon decision making. Experimental results demonstrate that the proposed information bottleneck-driven intrinsic motivation improves exploration efficiency, training stability, and solution quality compared to standard reinforcement learning baselines. Full article

The expansion of renewables in modern power systems and the coordinated development of upstream and downstream industrial chains are promoting a shift on the utility side from traditional settlement by energy toward operation driven by data and models. Industrial electricity consumption data exhibit pronounced multi-scale temporal structures and sectoral heterogeneity, which makes unified long-term load and generation forecasting while maintaining accuracy, interpretability, and scalability a challenge. From a modern system identification perspective, this paper proposes a System Identification in Adaptive Bayesian Framework (SIABF) for medium- and long-term industrial load forecasting based on daily freeze electricity time series. By combining daily aggregation of high-frequency data, frequency domain analysis, sparse identification, and long-term extrapolation, we first construct daily freeze series from 15 min measurements, and then we apply discrete Fourier transforms and a spectral complexity index to extract dominant periodic components and build an interpretable sinusoidal basis library. A sparse regression formulation with ${\ell }_1$ regularization is employed to select a compact set of key basis functions, yielding concise representations of sector and enterprise load profiles and naturally supporting multivariate and joint multi-sector modeling. Building on this structure, we implement a state-space-implicit physics-informed Bayesian forecasting model and evaluate it on real data from three representative sectors, namely, steel, photovoltaics, and chemical, using one year of 15 min measurements. Under a one-month-ahead evaluation using one year of 15 min measurements, the proposed framework attains a Mean Absolute Percentage Error (MAPE) of 4.5% for a representative PV-related customer case and achieves low single-digit MAPE for high-inertia sectors, often outperforming classical statistical models, sparse learning baselines, and deep learning architectures. These results should be interpreted as indicative given the limited time span and sample size, and broader multi-year, population-level validation is warranted. Full article

Remote sensing small object detection remains a challenging task due to limited feature representation and interference from complex backgrounds. Existing methods that rely exclusively on either visible or infrared modalities often fail to achieve both accuracy and robustness in detection. Effectively integrating cross-modal information to enhance detection performance remains a critical challenge. To address this issue, we propose a novel Multispectral Sparse Cross-Attention Guided Mamba Network (MSCGMN) for small object detection in remote sensing. The proposed MSCGMN architecture comprises three key components: Multispectral Sparse Cross-Attention Guidance Module (MSCAG), Dynamic Grouped Mamba Block (DGMB), and Gated Enhanced Attention Module (GEAM). Specifically, the MSCAG module selectively fuses RGB and infrared (IR) features using sparse cross-modal attention, effectively capturing complementary information across modalities while suppressing redundancy. The DGMB introduces a dynamic grouping strategy to improve the computational efficiency of Mamba, enabling effective global context modeling. In remote sensing images, small objects occupy limited areas, making it difficult to capture their critical features. We design the GEAM module to enhance both global and local feature representations for small object detection. Experiments on the VEDAI and DroneVehicle datasets show that MSCGMN achieves mAP50 scores of 83.9% and 84.4%, outperforming existing state-of-the-art methods and demonstrating strong competitiveness in small object detection tasks. Full article

Millimeter-wave (mmWave) massive multiple-input, multiple-output (MIMO) systems are a cornerstone technology for integrated sensing and communication (ISAC) in sixth-generation (6G) mobile networks. These systems provide high-capacity backhaul while simultaneously enabling high-resolution environmental sensing. However, accurate channel estimation remains highly challenging due to intrinsic noise sensitivity and clustered sparse multipath structures. These challenges are particularly severe under limited pilot resources and low signal-to-noise ratio (SNR) conditions. To address these difficulties, this paper proposes HASwinNet, a deep learning (DL) framework designed for mmWave channel denoising. The framework integrates a hierarchical Swin Transformer encoder for structured representation learning. It further incorporates two complementary branches. The first branch performs sparse token extraction guided by angular-domain significance. The second branch focuses on angular-domain refinement by applying discrete Fourier transform (DFT), squeeze-and-excitation (SE), and inverse DFT (IDFT) operations. This generates a mask that highlights angularly coherent features. A decoder combines the outputs of both branches with a residual projection from the input to yield refined channel estimates. Additionally, we introduce an angular-domain perceptual loss during training. This enforces spectral consistency and preserves clustered multipath structures. Simulation results based on the Saleh–Valenzuela (S–V) channel model demonstrate that HASwinNet achieves significant improvements in normalized mean squared error (NMSE) and bit error rate (BER). It consistently outperforms convolutional neural network (CNN), long short-term memory (LSTM), and U-Net baselines. Furthermore, experiments with reduced pilot symbols confirm that HASwinNet effectively exploits angular sparsity. The model retains a consistent advantage over baselines even under pilot-limited conditions. These findings validate the scalability of HASwinNet for practical 6G mmWave backhaul applications. They also highlight its potential in ISAC scenarios where accurate channel recovery supports both communication and sensing. Full article

In recent years, unmanned aerial vehicles (UAVs) have become increasingly prevalent across diverse application scenarios due to their high maneuverability, compact size, and cost-effectiveness. However, these advantages also introduce significant challenges for UAV detection in complex environments. This paper proposes an efficient feature perception network (EFPNet) for UAV detection, developed on the foundation of the RT-DETR framework. Specifically, a dual-branch HiLo-ConvMix attention (HCM-Attn) mechanism and a pyramid sparse feature transformer network (PSFT-Net) are introduced, along with the integration of a DySample dynamic upsampling module. The HCM-Attn module facilitates interaction between high- and low-frequency information, effectively suppressing background noise interference. The PSFT-Net is designed to leverage deep-level features to guide the encoding and fusion of shallow features, thereby enhancing the model’s capability to perceive UAV texture characteristics. Furthermore, the integrated DySample dynamic upsampling module ensures efficient reconstruction and restoration of feature representations. On the TIB and Drone-vs-Bird datasets, the proposed EFPNet achieves mAP50 scores of 94.1% and 98.1%, representing improvements of 3.2% and 1.9% over the baseline models, respectively. Our experimental results demonstrate the effectiveness of the proposed method for small UAV detection. Full article

Seafloor 3D reconstruction is a core technology for seafloor topography and deformation monitoring. Due to the complexity of the deep-sea environment and the high requirements for measurement devices, long-term monitoring can only acquire low-resolution and limited seafloor topography data. This leads to difficulties for existing 3D reconstruction algorithms in handling details and accuracy, especially with complex variations in seafloor terrain, which poses higher demands on 3D reconstruction algorithms. This study proposes a “fractal–Gaussian process” hybrid model, leveraging the fractal self-similarity property to precisely capture complex local details of the seafloor terrain, combined with the Bayesian global optimization ability of the Gaussian process model, to achieve high-resolution modeling of seafloor 3D reconstruction. Finally, Perlin noise is introduced to enhance the naturalness and detail representation of the terrain. Experiments show that under sparse data conditions, the proposed method significantly outperforms traditional interpolation methods, with average errors reduced by 30–40% and an R² value of 0.9836. Full article