We investigate the convergence properties of incremental mirror descent type subgradient algorithms for minimizing the sum of convex functions. In each step, we only evaluate the subgradient of a single component function and it back to the feasible domain, which makes iterations very cheap to compute. The analysis is made for a randomized selection of the component functions, which yields the deterministic algorithm as a special case. Under supplementary differentiability assumptions on the function which induces the mirror map, we are also able to deal with the presence of another term in the objective function, which is evaluated via a proximal type step. In both cases, we derive convergence rates of in expectation for the th best objective function value and illustrate our theoretical findings by numerical experiments in positron emission tomography and machine learning.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC6382287 | PMC |
| http://dx.doi.org/10.1080/02331934.2018.1482491 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Effects of Different Blood Flow Restriction Training Modes on Body Composition and Maximal Strength of Untrained Individuals.
Life (Basel)
December 2024
Exercise and Metabolism Research Center, College of Physical Education and Health Sciences, Zhejiang Normal University, Jinhua 321004, China.
Background: The objective of this study was to examine the impacts of absolute cuff pressure blood flow restriction (A-BFR) training and incremental cuff pressure blood flow restriction (I-BFR) training, under equal cuff pressures, on body composition and maximal strength among untrained adults. Additionally, we aimed to compare these effects with those observed in high-load resistance training (HL-RT).
Methods: Thirty-three adults without prior professional sports or resistance training experience were recruited and randomly assigned to three groups ( = 11 per group) for an 8-week training program, held three times weekly.
F-DPA714 PET/MRI as a potential imaging tool for detecting possible antibody-negative autoimmune encephalitis: a prospective study.
J Neurol
December 2024
Department of Neurology and Institute of Neurology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, 197 Ruijin 2nd Road, Shanghai, 200025, China.
Background And Objectives: Conventional magnetic resonance imaging (MRI) used for detecting possible antibody-negative autoimmune encephalitis (AIE) often fails to meet the diagnostic requirements of this disease. Positron emission tomography (PET) with a translocator protein radioligand can help visualize microglia distribution density in inflammation-related diseases, thereby offering potentially incremental value to conventional MRI for the in vivo assessment of possible antibody-negative AIE.
Methods: In this prospective study, 15 participants diagnosed with possible antibody-negative AIE and 10 healthy controls were enrolled (ClinicalTrials.
Investigations into the Material Characteristics of Selected Plastics Manufactured Using SLA-Type Additive Methods.
Polymers (Basel)
June 2024
Department of Mechanics, Materials and Biomedical Engineering, Faculty of Mechanical Engineering, Wroclaw University of Science and Technology, Smoluchowskiego 25 Str., 50-370 Wroclaw, Poland.
In this study, the authors performed a strength analysis of seven groups of commercially available materials based on SLS incremental technology. Test samples were made with Original PRUSA SL1S printers, with 10 samples of each type from 7 resins selected for testing. The tests were carried out on an MTS Bionix machine in a static tensile test, during which the basic mechanical properties were determined.
View Article and Find Full Text PDFNeuroLab 2.0: An Alternative Storyline Design Approach for Translating a Research-Based Summer Experience into an Advanced STEM+M Curriculum Unit that Supports Three-Dimensional Teaching and Learning in the Classroom.
J STEM Outreach
March 2024
Coastal Marine Biolabs Integrative Biosciences Program, Ventura Harbor, CA.
In this case study, we describe an alternative storyline design approach that we adopted to translate an informal, out-of-school summer science experience with a strong emphasis on developmental neuroscience and data literacy into a more inclusive, replicable, and scalable experience for formal high school science instruction. Combining elements of problem- and project-based learning, a storyline is a curriculum model that engages students in the application of investigative science and engineering practices to incrementally build conceptual models that explain an observable (anchoring) phenomenon. Published reports on the storyline design process describe procedures and tools that are well suited to the creation of novel instructional units.
View Article and Find Full Text PDFEvaluating the Long-Term Cost-Effectiveness of the English NHS Diabetes Prevention Programme using a Markov Model.
Pharmacoecon Open
July 2024
Health Organisation, Policy and Economics, Division of Population Health, Health Services Research and Primary Care, School of Health Sciences, Faculty of Biology, Medicine and Health, The University of Manchester, Suite 12, Floor 7, Williamson Building, Oxford Road, Manchester, M13 9PL, UK.
Background: In 2016, England launched the largest nationwide diabetes mellitus prevention programme, the NHS Diabetes Prevention Programme (NHS DPP). This paper seeks to evaluate the long-term cost-effectiveness of this programme.
Methods: A Markov cohort state transition model was developed with a 35-year time horizon and yearly cycles to compare referral to the NHS DPP to usual care for individuals with non-diabetic hyperglycaemia.
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!