Results 81 to 90 of about 28,341 (292)
Archaeological Prospection, EarlyView.ABSTRACT The detection of buried or obscured archaeological features remains a central challenge in landscape archaeology, particularly in the irrigated floodplains of Mesopotamia where levees and canals formed the basis of complex agrarian systems. This study presents a deep learning–based approach for the large‐scale, automated detection of ancient ...
Nazarij Buławka +4 more
wiley +1 more source
Deep Underground Science and Engineering, EarlyView.This paper presents temporal and adaptive‐frequency network with MixStyle (TAMNet), a deep time‐series modeling framework for accurate and robust multi‐well oil productivity forecasting. TAMNet integrates transformer and long short‐term memory architectures to capture both short‐ and long‐term temporal dependencies, enhanced by a temporal gate unit ...
Chunxi Yang +6 more
wiley +1 more source
Доклады Белорусского государственного университета информатики и радиоэлектроники, 2019 Parallel algorithm searching the maximum of the objective function using CUDA technology based on the modified method of dynamic programming is presented.
E. N. Seredin
doaj
Energy Science &Engineering, EarlyView.This graphical abstract illustrates a reproducible pipeline that combines gradient‐boosting‐based feature selection with a CNN–BiLSTM–Transformer model to forecast solar irradiance across multi‐site satellite and ground datasets, delivering robust, high‐accuracy predictions that support sustainable grid planning and reliable PV integration.
Muhammad Farhan Hanif +5 more
wiley +1 more source
The technology of CUDA - implementation of non-graphic computing on GPGPU [PDF]
, 2014 Technologies of GPGPU are an important and promising area of high- performance parallel computing. This paper deals with one of the most popular technologies today. It is CUDA.
Парубочий, Віталій Олегович +1 more
core
Journal of Magnetic Resonance Imaging, EarlyView.ABSTRACT Background The demand for cardiac MRI is increasing with the growing burden of cardiovascular disease. However, conventional protocols require sequential acquisitions for multi‐breath‐hold 2D cine and 3D MR angiography (MRA), which is time‐consuming.
Ruixin Chen +7 more
wiley +1 more source
Parallelising wavefront applications on general-purpose GPU devices [PDF]
, 2010 Pipelined wavefront applications form a large portion of the high performance scientific computing workloads at supercomputing centres. This paper investigates the viability of graphics processing units (GPUs) for the acceleration of these codes, using ...
Pennycook, Simon J. +3 more
core
Journal of Magnetic Resonance Imaging, EarlyView.ABSTRACT Background The diffusion tensor image analysis along the perivascular space (DTI‐ALPS) index is an established term, and ongoing research aims to explore underlying mechanisms. Hypothesis DTI‐ALPS measures are influenced by subject‐specific orientational differences of the projection and association fibers. Study Type Prospective.
Dávid Bognár +9 more
wiley +1 more source
CUDA-LLM: LLMs Can Write Efficient CUDA Kernels
CoRRLarge Language Models (LLMs) have demonstrated strong capabilities in general-purpose code generation. However, generating the code which is deeply hardware-specific, architecture-aware, and performance-critical, especially for massively parallel GPUs, remains a complex challenge.
Wentao Chen +4 more
openaire +2 more sources
Predicting Statistical Signatures of Collective Emission in Disordered Color Center Ensembles
Laser &Photonics Reviews, EarlyView.We present an efficient simulation framework for modeling collective emission in disordered ensembles of quantum emitters. The low computational complexity enables large‐scale Monte Carlo simulations. Applied to SiV−${\rm SiV}^{-}$ clusters, it predicts thresholded superradiant bursts set by emitter number and quantum efficiency, as well as interaction‐
Qingyi Zhou +4 more
wiley +1 more source