Jove
Visualize
Contáctanos
JoVE
x logofacebook logolinkedin logoyoutube logo
ACERCA DE JoVE
Visión GeneralLiderazgoBlogCentro de Ayuda JoVE
AUTORES
Proceso de PublicaciónConsejo EditorialAlcance y PolíticasRevisión por ParesPreguntas FrecuentesEnviar
BIBLIOTECARIOS
TestimoniosSuscripcionesAccesoRecursosConsejo Asesor de BibliotecasPreguntas Frecuentes
INVESTIGACIÓN
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchivo
EDUCACIÓN
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualCentro de Recursos para ProfesoresSitio de Profesores
Términos y Condiciones de Uso
Política de Privacidad
Políticas

Videos de Conceptos Relacionados

Random Sampling Method01:09

Random Sampling Method

12.3K
Sampling is a technique to select a portion (or subset) of the larger population and study that portion (the sample) to gain information about the population. Data are the result of sampling from a population. The sampling method ensures that samples are drawn without bias and accurately represent the population. Because measuring the entire population in a study is not practical, researchers use samples to represent the population of interest. Among the various sampling methods used by...
12.3K
Sampling Distribution01:12

Sampling Distribution

13.6K
Given simple random samples of size n from a given population with a measured characteristic such as mean, proportion, or standard deviation for each sample, the probability distribution of all the measured characteristics is called a sampling distribution. How much the statistic varies from one sample to another is known as the sampling variability of a statistic. You typically measure the sampling variability of a statistic by its standard error. The standard error of the mean is an example...
13.6K
Stratified Sampling Method01:16

Stratified Sampling Method

12.8K
Sampling is a technique to select a portion (or subset) of the larger population and study that portion (the sample) to gain information about the population. The sampling method ensures that samples are drawn without bias and accurately represent the population. Because measuring the entire population in a study is not practical, researchers use samples to represent the population of interest.
To choose a stratified sample, divide the population into groups called strata and then take a...
12.8K
Cluster Sampling Method01:20

Cluster Sampling Method

12.7K
Appropriate sampling methods ensure that samples are drawn without bias and accurately represent the population. Because measuring the entire population in a study is not practical, researchers use samples to represent the population of interest.
To choose a cluster sample, divide the population into clusters (groups) and then randomly select some of the clusters. All the members from these clusters are in the cluster sample. For example, if you randomly sample four departments from your...
12.7K
Sampling Methods: Overview01:06

Sampling Methods: Overview

498
A sample refers to a smaller subset representative of a larger population. In analytical chemistry, studying or analyzing an entire population is often impractical or impossible. Therefore, samples are used to draw inferences and generalize the whole population. The sampling method selects individuals or items from a population to create a sample. Standard sampling methods include random, judgemental, systematic, stratified, and cluster sampling. 
In analytical chemistry, the choice of...
498
Sampling Methods: Sample Types01:18

Sampling Methods: Sample Types

405
Sampling materials are classified into three main types: solid, liquid, and gas.
Solid samples include a variety of substances, such as sediments from water bodies, soil, metals, and biological tissues. Two standard methods for extracting sediments from water bodies are grab sampling and piston coring. Grab sampling involves using a device to collect a discrete sediment sample from the bottom of a water body with minimal disturbance. Grab samples do not always represent the entire area due to...
405

También podría leer

Artículos Relacionados

Artículos vinculados a este trabajo por autores compartidos, revista y gráfico de citas.

Ordenar por
Same author

The genome sequence of <i>Bacillus anthracis</i> strain "Bundesheer" isolated from a postal sack in the US embassy Vienna in 2001 during the Amerithrax incident in the US.

Microbiology resource announcements·2026
Same author

Synthesizing vocal tract magnetic resonance imaging sequences with phoneme-aware diffusion models.

Journal of medical imaging (Bellingham, Wash.)·2026
Same author

Filter2Noise: a framework for interpretable and zero-shot low-dose CT image denoising.

Journal of medical imaging (Bellingham, Wash.)·2026
Same author

Multi-site distributed training with data protections for PET-based synthetic CT.

EJNMMI physics·2026
Same author

Challenging Hounsfield Unit cutoffs: spectral thresholding for synthetic coronary plaque phantoms on photon-counting CT.

Journal of medical imaging (Bellingham, Wash.)·2026
Same author

Vision transformer-based diagnosis of psoriasis and eczema in whole-slide histology.

Virchows Archiv : an international journal of pathology·2026

Video Experimental Relacionado

Updated: Sep 9, 2025

Sampling Soils in a Heterogeneous Research Plot
07:11

Sampling Soils in a Heterogeneous Research Plot

Published on: January 7, 2019

34.9K

Una técnica de muestreo robusta para la simulación de distribución realista en el aprendizaje federado

Robin Hoepp1,2, Leonhard Rist3,4, Alexander Katzmann3

  • 1Computed Tomography, Siemens Healthineers, Forchheim, Germany. robin.hoepp@fau.de.

International journal of computer assisted radiology and surgery
|September 2, 2025
PubMed
Resumen
Este resumen es generado por máquina.

La formación de aprendizaje federado (FL) puede verse perjudicada por las distribuciones de datos no identificados. Un nuevo algoritmo de muestreo simula distribuciones de etiquetas realistas para analizar la degradación del rendimiento de FL antes del despliegue.

Palabras clave:
Cambio de distribuciónAprendizaje federadoMuestreo

Más Videos Relacionados

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

681
An Unbiased Approach of Sampling TEM Sections in Neuroscience
10:56

An Unbiased Approach of Sampling TEM Sections in Neuroscience

Published on: April 13, 2019

7.3K

Videos de Experimentos Relacionados

Last Updated: Sep 9, 2025

Sampling Soils in a Heterogeneous Research Plot
07:11

Sampling Soils in a Heterogeneous Research Plot

Published on: January 7, 2019

34.9K
Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
03:14

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

681
An Unbiased Approach of Sampling TEM Sections in Neuroscience
10:56

An Unbiased Approach of Sampling TEM Sections in Neuroscience

Published on: April 13, 2019

7.3K

Área de la Ciencia:

  • Aprendizaje automático
  • Inteligencia artificial
  • La informática médica

Sus antecedentes:

  • El aprendizaje federado (FL) permite entrenar modelos de aprendizaje profundo en datos descentralizados, cruciales para entornos clínicos sensibles a la privacidad.
  • Los datos no independientes y distribuidos de manera idéntica (no IID), que surgen de las variaciones demográficas entre los clientes, pueden degradar significativamente el rendimiento del modelo FL.
  • La evaluación del impacto de las distribuciones de datos no IDI es vital antes de implementar FL a gran escala en el cuidado de la salud.

Objetivo del estudio:

  • Desarrollar y evaluar un nuevo algoritmo de muestreo para crear distribuciones de etiquetas realistas y orientadas al cliente.
  • Investigar la degradación del rendimiento de los modelos FL en escenarios simulados de datos no IID.
  • Proporcionar un método eficiente para analizar los efectos de la heterogeneidad de los datos en FL.

Principales métodos:

  • Se desarrolló un algoritmo de muestreo para generar subconjuntos de datos con medias y desviaciones estándar especificadas de una distribución global.
  • Las medidas de impureza Chi-cuadrado y Gini se emplearon para la optimización numérica de las distribuciones de etiquetas en múltiples grupos.
  • El algoritmo se aplicó a un conjunto de datos clínicos del mundo real para la estimación de peso y altura basada en cámaras 3D.

Principales resultados:

  • El entrenamiento Federated Averaging (FedAvg) con datos no IDI muestreados dio lugar a una caída en el rendimiento.
  • En el modelo global se observó un deterioro realista del 25,3% para las estimaciones de peso y del 28,7% para las estimaciones de altura.
  • La técnica de muestreo propuesta demostró un impacto negativo significativo en comparación con una línea de base dividida de datos concretos.

Conclusiones:

  • Las distribuciones de etiquetas sesgadas por el cliente en los entornos de FL pueden perjudicar sustancialmente el entrenamiento y el rendimiento del modelo.
  • El algoritmo de muestreo desarrollado ofrece un enfoque eficiente para el análisis previo al despliegue de los efectos de los datos no IDI.
  • Esta técnica es versátil, aplicable a varias arquitecturas de red, escenarios clínicos y subpoblaciones no IDI.