-
Article
Open AccessEvaluation of deep learning-based reconstruction late gadolinium enhancement images for identifying patients with clinically unrecognized myocardial infarction
The presence of infarction in patients with unrecognized myocardial infarction (UMI) is a critical feature in predicting adverse cardiac events. This study aimed to compare the detection rate of UMI using conv...
-
Article
Open AccessAbove-room-temperature chiral skyrmion lattice and Dzyaloshinskii–Moriya interaction in a van der Waals ferromagnet Fe3−xGaTe2
Skyrmions in existing 2D van der Waals (vdW) materials have primarily been limited to cryogenic temperatures, and the underlying physical mechanism of the Dzyaloshinskii–Moriya interaction (DMI), a crucial ing...
-
Article
How good are machine learning clouds? Benchmarking two snapshots over 5 years
We conduct an empirical study of machine learning functionalities provided by major cloud service providers, which we call machine learning clouds. Machine learning clouds hold the promise of hiding all the sophi...
-
Article
Open AccessStochastic gradient descent without full data shuffle: with applications to in-database machine learning and deep learning systems
Modern machine learning (ML) systems commonly use stochastic gradient descent (SGD) to train ML models. However, SGD relies on random data order to converge, which usually requires a full data shuffle. For in-...
-
Article
A binary packing material-based method for estimating small-strain shear modulus of sandy soils
A set of creative formulation is introduced to estimating the small-strain shear modulus (Gmax) of various sandy soils through the measurement of well-controlled laboratory tests combined with the analysis of exp...
-
Article
Improved laser-induced dynamic wetting behavior and spreading mechanism of non-reactive Mg/steel systems with the assistance of Cu coating
A Cu coating was introduced on the steel surface by electroplating, and the wetting and spreading behavior of Mg alloys on Cu-coated steel substrates during the laser machining process was investigated. Wettin...
-
Article
A systematic evaluation of machine learning on serverless infrastructure
Recently, the serverless paradigm of computing has inspired research on its applicability to data-intensive tasks such as ETL, database query processing, and machine learning (ML) model training. Recent efforts h...
-
Article
Open AccessSpatial patterns of Holocene temperature changes over mid-latitude Eurasia
The Holocene temperature conundrum, the discrepancy between proxy-based Holocene global cooling and simulated global annual warming trends, remains controversial. Meanwhile, reconstructions and simulations sho...
-
Article
Open AccessAn energy-free strategy to elevate anti-icing performance of superhydrophobic materials through interfacial airflow manipulation
Superhydrophobic surfaces demonstrate excellent anti-icing performance under static conditions. However, they show a marked decrease in icing time under real flight conditions. Here we develop an anti-icing st...
-
Chapter and Conference Paper
Optimization of Non-destructive Detection Method for Metal Pipelines Based on Magnetic Induction Tomography
Pipeline transportation has become an indispensable way in production and life. However, the emergence of pipeline defects causes great harm, so timely detection of pipeline defects is crucial for all fields. ...
-
Article
Feature Extraction and Intelligent Fault Diagnosis of Marine Machinery
The present work proposes a new method to realize the intelligent condition monitoring and fault diagnosis of marine machinery.
-
Article
Effect of CoCrNi Interlayer on the Microstructure, Mechanical, and Corrosion Properties of Cold Metal Transfer Al/Steel Joints
Cold metal transfer (CMT) welding is an advanced arc welding technology, which is promising for joining dissimilar materials such as aluminum, magnesium, titanium alloys and steels for automotive and aerospace...
-
Article
Open AccessContribution of boundary non-stoichiometry to the lower-temperature plasticity in high-pressure sintered boron carbide
The improvement of non-oxide ceramic plasticity while maintaining the high-temperature strength is a great challenge through the classical strategy, which generally includes decreasing grain size to several na...
-
Article
Effect of bending radius on deformation behavior of H62 brass tubes in a less constrained free bending process
Finite element method and bending experiments were carried out to survey the influence of bending radius on forming quality of H62 brass tubes in free bending process，which are widely used as key components of...
-
Article
Tensor completion via hybrid shallow-and-deep priors
For the ill-posed tensor completion (TC) problem, how to appropriately explore the hidden information is a pivotal yet challenging issue. Two well-known priors have been intensively studied, including the shal...
-
Article
LncRNA RMRP Contributes to the Development and Progression of Spinal Cord Injury by Regulating miR-766-5p/FAM83A Axis
Spinal cord injury (SCI) is known as a central nervous system disorder. Previous studies suggested that long-non-coding RNA RMRP (LncRNA RMRP) was abnormally expressed in SCI, but the potential underlying mech...
-
Article
Upregulation of P2Y12 inhibits chondrocyte apoptosis in lumbar osteoarthritis through the PI3K/AKT signaling pathway
Lumbar facet osteoarthritis (FJOA) is a major cause of severe lower back pain and disability worldwide. However, the mechanism underlying cartilage degeneration in FJOA remains unclear. The purpose of this stu...
-
Chapter
Introduction
Machine learning (ML) has shown its ability to extract buried knowledge from data that is hard for humans to understand. To obtain a satisfying machine learning model, we often use some optimization algorithms...
-
Chapter
Conclusion
Machine learning has been the driving force of many real-world applications. Many machine learning models rely on gradient optimization algorithms that fit the model parameters over the training data. As the d...
-
Chapter
Basics of Distributed Machine Learning
Extending the gradient optimization of a machine learning model, which is originally serial, to a distributed setting is nontrivial. Several fundamental techniques are involved in meeting the characteristics o...
