A5080057983- Sparse and Compressive Sensing Techniques
- Advanced Bandit Algorithms Research
- Stochastic Gradient Optimization Techniques
- Big Data and Business Intelligence
- Technology Adoption and User Behaviour
- Optimization and Search Problems
- Tensor decomposition and applications
- RNA regulation and disease
- Statistical Methods and Inference
- Aortic Disease and Treatment Approaches
- Probabilistic and Robust Engineering Design
- Cardiac Imaging and Diagnostics
- Aortic aneurysm repair treatments
- Brain Tumor Detection and Classification
- Neural Networks and Applications
- Machine Learning and Algorithms
- Machine Learning and Data Classification
- Advanced X-ray and CT Imaging
- Web visibility and informetrics
- RNA modifications and cancer
- Distributed Sensor Networks and Detection Algorithms
- Infectious Aortic and Vascular Conditions
- Cancer-related molecular mechanisms research
Qingdao University
2022-2024
Qingdao Municipal Hospital
2024
Affiliated Hospital of Qingdao University
2022-2024
Peking University
2024
Type A aortic dissection (TAAD) has a rapid onset and high mortality. Currently, diameter is the major criterion for evaluating risk of TAAD. We attempted to find other morphological indicators further analyze their relationships with type dissection.We included imaging clinical data 112 patients. The patients were divided into three groups, which Group 1 had 49 normal diameter, 2 22 ascending aneurysm, 3 41 used AW Server software, version 3.2, measure aorta-related indicators.First, in 1,...
Introduction. Governance process optimisation is critical to achieve the goal of improving public services efficiency. Public data service crucial starting point for realising this goal. However, practical challenges persist in service, including unclear processes and insufficient identification citizens’ information needs. Method & Analysis. We employed an inductive methodology analyse users’ needs annual distribution among Based on a data-driven approach, we mapped these sub-categories...
Abstract Acute myocardial infarction (AMI) is a critical cardiovascular disease with significant health implications. This study aims to investigate the role of RNA Modification-Related Genes (RMRGs), which are essential post-transcriptional regulators, in pathology AMI. By examining AMI-related datasets (GSE24519, GSE48060, GSE34198), RMRGs were collected from GeneCards and PubMed. The analysis involved enrichment analyses using Gene Expression Omnibus (GO), Kyoto Encyclopedia Genomes...
This paper presents a simple yet efficient method for statistical inference of tensor linear forms with incomplete and noisy observations. Under the Tucker low-rank model, we utilize an appropriate initial estimate, along debiasing technique followed by one-step power iteration, to construct asymptotic normal test statistic. is suitable various tasks, including confidence interval prediction, under heteroskedastic sub-exponential noises, simultaneous testing. Furthermore, approach reaches...
This paper investigates regret minimization, statistical inference, and their interplay in high-dimensional online decision-making based on the sparse linear context bandit model. We integrate $\varepsilon$-greedy algorithm for with a hard thresholding estimating parameters introduce an inference framework debiasing method using inverse propensity weighting. Under margin condition, our achieves either $O(T^{1/2})$ or classical $O(T^{1/2})$-consistent indicating unavoidable trade-off between...
We study the contextual bandits with knapsack (CBwK) problem under high-dimensional setting where dimension of feature is large. The reward pulling each arm equals multiplication a sparse weight vector and current arrival, additional random noise. In this paper, we investigate how to exploit sparsity structure achieve improved regret for CBwK problem. To end, first develop an online variant hard thresholding algorithm that performs estimation in manner. further combine our estimator...
Many important tasks of large-scale recommender systems can be naturally cast as testing multiple linear forms for noisy matrix completion. These problems, however, present unique challenges because the subtle bias-and-variance tradeoff and an intricate dependence among estimated entries induced by low-rank structure. In this paper, we develop a general approach to overcome these difficulties introducing new statistics individual tests with sharp asymptotics both marginally jointly,...
This paper introduces a dual-based algorithm framework for solving the regularized online resource allocation problems, which have potentially non-concave cumulative rewards, hard constraints, and non-separable regularizer. Under strategy of adaptively updating proposed only requests approximate solutions to empirical dual problems up certain accuracy yet delivers an optimal logarithmic regret under locally second-order growth condition. Surprisingly, delicate analysis objective function...