67639 research outputs found
Sort by
A multi-scale feature extraction and fusion framework based on wavelet Kolmogorov-Arnold networks and parallel Bi-directional gated recurrent units for electric load forecasting
No full textShort-term electric load forecasting remains challenged by the dual requirements of accuracy and robustness due to the combined effects of strong seasonality, multi-scale spikes, and stochastic disturbances. To address this, we propose a novel multi-scale forecasting framework, NP-WavKAN-Fusion, which integrates Neural Prophet for data decomposition and a Wavelet-based Kolmogorov–Arnold Network (WavKAN) with learnable wavelet kernels for multi-scale encoding. This fusion model utilizes a Bi-directional Gated Recurrent Unit (BiGRU) to capture long-term temporal dependencies and an adaptive feature fusion gate (AFF) to dynamically re-weight static and dynamic features for final load predictions. Extensive experiments on two public datasets from Australia and Morocco show that NP-WavKAN-Fusion consistently outperforms traditional models, reducing the mean absolute error by at least 30 %. For multi-step forecasting tasks, NP-WavKAN-Fusion maintains error inflation within 15 %, demonstrating superior performance compared to state-of-the-art long-sequence models such as Informer and PatchTST. The Diebold–Mariano test confirms that NP-WavKAN-Fusion yields statistically significant improvements, with 19 out of 20 comparisons showing lower errors. Ablation studies show that removing either the Neural Prophet component or the AFF significantly increases the forecasting error, validating the necessity of our layered denoising and fusion strategies. The proposed NP-WavKAN-Fusion framework demonstrates strong potential for real-world applications in electric load forecasting, offering robust performance under various temporal and non-stationary conditions.This work was supported by the National Natural Science Foundation of China (Grant No. 52406267) and the Major Science and Technology Project of Xinjiang Autonomous Region, China (Grant No. 2022A01004-4)
CCDC 2362699: Experimental Crystal Structure Determination : bis[6-(1,3-dioxo-1,3-dihydro-2H-benzo[f]isoindol-2-yl)hexan-1-aminium] hexaiodo-tin(iv)
No full textAn entry from the Cambridge Structural Database, the world’s repository for small molecule crystal structures. The entry contains experimental data from a crystal diffraction study. The deposited dataset for this entry is freely available from the CCDC and typically includes 3D coordinates, cell parameters, space group, experimental conditions and quality measures
Diversity, Functionality and Adaptations of Benthic Microbial Communities in Tropical Habitats using a Metagenomic Approach
No full textMicrobial life forms the unseen foundation of marine ecosystems, playing essential roles in biogeochemical cycling, ecosystem functioning, and the regulation of anthropogenic pressures. Yet, benthic microbial communities remain among the least characterized components of the ocean microbiome. This thesis uses shotgun metagenomics to explore the taxonomic composition and functional potential of benthic microbial assemblages in tropical marine habitats, with a focus on global ocean patterns, the Red Sea, and the Bahamas.
First, the construction of the KAUST Metagenomic Analysis Platform (KMAP) Ocean Gene Catalog 1.0, the largest marine gene catalog to publication date, brought a baseline inventory of microbial diversity and gene function across different oceanic realms, while underscoring the reduced number of studies and spatial-scale underrepresentation of benthic habitats. In order to fill this gap, we studied the domain-level structure of the Red Sea benthic habitats, analyzing taxonomic distribution across the Web of Life in this understudied context. Increasing anthropogenic pressures, from coastal urbanization to the influx of xenobiotics, have significantly reshaped benthic ecosystems, even along the once-pristine eastern coast of the Red Sea. Antibiotic resistance and plastic pollution have emerged as emblematic indicators of Anthropocene-driven change, revealing the pervasive human imprint on benthic microbial communities. This work examines the diversity and spatial distribution of plastic polymer-degrading enzymes as well as the abundance and biogeographic structuring of antibiotic resistance genes (ARGs), highlighting microbial adaptation and tracing pollution exposure across spatial gradients and habitat types. To contextualize these patterns in a broader framework, Bahamian seagrass sediments were analyzed as a parallel system. Situated at similar latitudes to the central Red Sea and hosting the world’s largest seagrass meadow, the Bahamas offer a comparable setting. Microbial community composition and nutrient-cycling traits reveal ecosystem-level signatures of microbial adaptation in shallow, vegetated benthic habitats. Together, these chapters provide a picture of different aspects of benthic microbial life across scales, from global oceans to regional ecosystems. The findings offer critical insights into the microbial role in environmental resilience, pollutant degradation and ecosystem service provisioning in tropical benthic habitats, reinforcing the need to integrate benthic microbiomes into marine biodiversity frameworks and conservation agendas
THz-Band Near-Field RIS Channel Modeling for Linear Channel Estimation
No full textReconfigurable intelligent surface (RIS)-aided terahertz (THz)-band communications are promising enablers for future wireless networks. However, array densification at high frequencies introduces significant challenges in accurate channel modeling and estimation, particularly with THz-specific fading, mutual coupling (MC), spatial correlation, and near-field effects. In this work, we model THz outdoor small-scale fading channels using the mixture gamma (MG) distribution, considering absorption losses, spherical wave propagation, MC, and spatial correlation across large base stations and RISs. We derive the distribution of the cascaded RIS-aided channel and investigate linear channel estimation techniques, analyzing the impact of various channel parameters. Numerical results based on precise THz parameters reveal that accounting for spatial correlation, MC, and near-field modeling substantially enhances estimation accuracy, especially in ultra-massive arrays and short-range scenarios. These results underscore the importance of incorporating these effects for precise, physically consistent channel modeling
When Every Total Program Is a Finite Tree-Program: A Study of Program-Saturated Classes of Structures
No full textThis paper investigates classes of structures and individual structures where programs implementing functions defined everywhere (total programs) are equivalent to finite tree-programs. The programs considered may include cycles and contain at most countably many nodes. The analysis begins with programs where arbitrary terms of a given signature are used in function nodes, and arbitrary formulas of this signature are used in predicate nodes. The results are then extended to programs that closely resemble computation trees: if such a program is a finite tree-program, it can be classified as an ordinary computation tree.The research reported in this publication was supported by King Abdullah University of Science and Technology (KAUST).The author is greatly indebted to the anonymous reviewers for useful comments and suggestions
From Density to Void: Why Brain Networks Fail to Reveal Complex Higher-Order Structures
No full textIn brain network analysis using resting-state fMRI, there is growing interest in modeling higher-order interactions beyond simple pairwise connectivity via persistent homology. Despite the promise of these advanced topological tools, robust and consistently observed higher-order interactions over time remain elusive. In this study, we investigate why conventional analyses often fail to reveal complex higher-order structures - such as interactions involving four or more nodes - and explore whether such interactions truly exist in functional brain networks. We utilize a simplicial complex framework often used in persistent homology to address this question.NIH grants EB028753, MH133614 and NSF grant MDS-201077. We would like to thank D. Vijay Anand of University College London for generating Figure 1 that was originally published in [2]
Scalable Low-Rank Solvers for Large-Scale 3D Mesh Deformation Using Global and Compact Support RBF Kernels
No full textRadial Basis Function (RBF) interpolation is a widely used technique for maintaining high-quality 3D unstructured meshes in fluid-structure interaction, ensuring smooth deformation while preserving mesh integrity. This paper studies the rank distribution of global and compact support RBF kernels using Tile Low-Rank (TLR) Cholesky solver, focusing on their impact on computational efficiency and scalability. We analyze the arithmetic intensity and performance of these RBFs through the hierarchical Roofline model to understand computational trade-offs. Large-scale experiments are conducted to evaluate the scalability of TLR Cholesky solver for two kernels: Gaussian (global support) and Wendland (compact support). As a real-world application, we examine the SARS-CoV-2 viral mesh, where RBF-based deformation is employed to model viral populations in dynamic fluid environments. We adopt a sphere-packing approach to arrange viruses in a dense, non-overlapping configuration, modeling real-world constraints. Our findings show that Gaussian's lower-rank intensity hinders scalability, whereas Wendland's compact support results in higher computational cost. The results demonstrate that the TLR approach reveals performance characteristics that dense factorization cannot, offering insights into kernel selection, scalability, and computational trade-offs for large-scale RBF-based mesh deformation.United States Department of Defense Defense Advanced Research Projects Agency (DARPA
Tinyml Nlp Scheme for Semantic Wireless Sentiment Classification with Privacy Preservation
No full textNatural Language Processing (NLP) operations, such as semantic sentiment analysis and text synthesis, often raise privacy concerns and demand significant on-device computational resources. Centralized learning (CL) on the edge provides an energy-efficient alternative but requires collecting raw data, compromising user privacy. While federated learning (FL) enhances privacy, it imposes high computational energy demands on resource-constrained devices. This study provides insights into deploying privacy-preserving, energy-efficient NLP models on edge devices. We introduce semantic split learning (SL) as an energy-efficient, privacy-preserving tiny machine learning (TinyML) framework and compare it to FL and CL in the presence of Rayleigh fading and additive noise. Our results show that SL significantly reduces computational power and CO2 emissions while enhancing privacy, as evidenced by a fourfold increase in reconstruction error compared to FL and nearly eighteen times that of CL. In contrast, FL offers a balanced trade-off between privacy and efficiency. Our code is available for replication at our GitHub repository: https://github.com/AhmedRadwan02/TinyEco2AI-NLP.This work is supported by the KAUST Office of Sponsored Research under Award ORA-CRG2021-4695
Role of desalinated seawater groundwater recharge for seasonal energy storage
No full textThis study examines the potential of using desalinated seawater for seasonal groundwater recharge as an innovative energy storage strategy in arid regions, focusing on Qatar. It proposes an alternative of seasonal storage to optimize energy and water supply by recharging aquifers with desalinated seawater during low electricity demand, primarily in winter. The study evaluates the groundwater recharge potential, identifying suitable sites based on hydrogeological parameters and assessing the energy storage capacity for optimized system efficiency. Results show that up to 2.57 TWh of seasonal energy can be stored using groundwater recharge, equivalent to 40.6 % of Qatar's seasonal energy storage needs, assuming a constant monthly supply of electricity and increased in water demand for agriculture. Additionally, storing desalinated water underground offers added benefits such as thermal energy storage, improved water security, and reduced groundwater salinity. The paper proposes two reservoir locations to store energy and water in Qatar. It provides a detailed cost assessment, concluding that desalinated groundwater recharge presents a cost-effective (estimated energy storage investment cost of 0.385 USD/kWh), scalable, and multi-functional solution to support Qatar's energy and water security objectives as it pursues decarbonization targets under Vision 2030