Random features approach has been widely used for kernel approximation in large-scale machine learning. A number of recent studies have explored data-dependent sampling of features, modifying the stochastic oracle from which random features are sampled. While proposed techniques in this realm improve the approximation, their suitability is often verified on a single learning task. In this article, we propose a task-specific scoring rule for selecting random features, which can be employed for different applications with some adjustments. We restrict our attention to canonical correlation analysis (CCA) and provide a novel, principled guide for finding the score function maximizing the canonical correlations. We prove that this method, called optimal randomized CCA (ORCCA), can outperform (in expectation) the corresponding kernel CCA with a default kernel. Numerical experiments verify that ORCCA is significantly superior to other approximation techniques in the CCA task.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/TNNLS.2021.3124868 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Effect of digital health applications with or without gamification on physical activity and cardiometabolic risk factors: a systematic review and meta-analysis of randomized controlled trials.
EClinicalMedicine
October 2024
Toronto 3D Knowledge Synthesis and Clinical Trials Unit, Clinical Nutrition and Risk Factor Modification Center, St. Michael's Hospital, Unity Health Toronto, Toronto, ON M5B 1W8, Canada.
Background: Use of health applications (apps) to support healthy lifestyles has intensified. Different app features may support effectiveness, including gamification defined as the use of game elements in a non-game situation. Whether health apps with gamification can impact behaviour change and cardiometabolic risk factors remains unknown.
View Article and Find Full Text PDFEndovascular treatment of postoperative aortic coarctation aneurysms-a single center experience.
Front Cardiovasc Med
December 2024
Department of Cardiology, University Hospital 'St. Ekaterina', Medical University of Sofia, Sofia, Bulgaria.
Background: Formation of local type aortic aneurysm years after surgical repair of coarctation (CoA) occurs in 10% of patients independent of the surgical technique and is a potentially life-threatening condition if left untreated with a high risk of aortic rupture. Redo open surgery is associated with 14% in-hospital mortality and a high risk of complications. Endovascular treatment appears to be a feasible alternative with a high success rate and low morbidity and mortality, but data concerning long-term results is still mandatory.
View Article and Find Full Text PDFDraw+: network-based computational drug repositioning with attention walking and noise filtering.
Health Inf Sci Syst
December 2025
Division of Software, Yonsei University, Mirae Campus, Yeonsedae-gil 1, Wonju-si, 26493 Gangwon-do Korea.
Purpose: Drug repositioning, a strategy that repurposes already-approved drugs for novel therapeutic applications, provides a faster and more cost-effective alternative to traditional drug discovery. Network-based models have been adopted by many computational methodologies, especially those that use graph neural networks to predict drug-disease associations. However, these techniques frequently overlook the quality of the input network, which is a critical factor for achieving accurate predictions.
View Article and Find Full Text PDFMulti-modal prediction of extracorporeal support-a resource intensive therapy, utilizing a large national database.
JAMIA Open
February 2025
Artificial Intelligence (AI) for Health Institute (AIHealth), Washington University in St Louis, St Louis, MO 63130, United States.
Objective: Extracorporeal membrane oxygenation (ECMO) is among the most resource-intensive therapies in critical care. The COVID-19 pandemic highlighted the lack of ECMO resource allocation tools. We aimed to develop a continuous ECMO risk prediction model to enhance patient triage and resource allocation.
View Article and Find Full Text PDFA novel multi-level 3D pose estimation framework for gait detection of Parkinson's disease using monocular video.
Front Bioeng Biotechnol
December 2024
Department of Rehabilitation Medicine, University of Hong Kong-Shenzhen Hospital, Shenzhen, China.
Introduction: Parkinson's disease (PD) is characterized by muscle stiffness, bradykinesia, and balance disorders, significantly impairing the quality of life for affected patients. While motion pose estimation and gait analysis can aid in early diagnosis and timely intervention, clinical practice currently lacks objective and accurate tools for gait analysis.
Methods: This study proposes a multi-level 3D pose estimation framework for PD patients, integrating monocular video with Transformer and Graph Convolutional Network (GCN) techniques.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!