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

Gradient and Del Operator01:14

Gradient and Del Operator

4.3K
In mathematics and physics, the gradient and del operator are fundamental concepts used to describe the behavior of functions and fields in space. The gradient is a mathematical operator that gives both the magnitude and direction of the maximum spatial rate of change. Consider a person standing on a mountain. The slope of the mountain at any given point is not defined unless it is quantified in a particular direction. For this reason, a "directional derivative" is defined, which is a vector...
4.3K
Survival Tree01:19

Survival Tree

369
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
 Building a Survival Tree
Constructing a...
369
Avoidance Learning and Learned Helplessness01:14

Avoidance Learning and Learned Helplessness

2.5K
Avoidance learning and learned helplessness are critical concepts in understanding behavioral responses to negative stimuli.
Avoidance learning occurs when an organism learns that a specific behavior can prevent an unpleasant outcome. For example, a student who receives a bad grade may start studying harder to avoid future poor grades. This behavior persists even when the negative outcome is no longer present. Avoidance learning is powerful because it maintains behavior in the absence of the...
2.5K
Introduction to Learning01:18

Introduction to Learning

884
Learning is the process of acquiring knowledge or skills through practice or experience, leading to long-lasting behavioral changes. This acquisition occurs through interaction with the environment and requires practice or experience. For instance, mastering a skill such as surfing requires considerable practice and experience, highlighting the essential role of repeated interactions with the environment in learning.
In contrast to learned behaviors, unlearned behaviors such as crying, sexual...
884
Forced Transdifferentiation01:28

Forced Transdifferentiation

2.3K
Transdifferentiation, also known as lineage reprogramming, was first discovered by Selman and Kafatos in 1974 in silkmoths. They observed that the moths’ cuticle-producing cells transformed into salt-producing cells. Many such cases of natural transdifferentiation occur in organisms. In humans, pancreatic alpha cells can become beta cells. In newts, the loss of the eye’s lens causes the pigmented epithelial cells to transdifferentiate into the lens cells.
Artificial...
2.3K
What is an Electrochemical Gradient?01:26

What is an Electrochemical Gradient?

126.4K
Adenosine triphosphate, or ATP, is considered the primary energy source in cells. However, energy can also be stored in the electrochemical gradient of an ion across the plasma membrane, which is determined by two factors: its chemical and electrical gradients.
The chemical gradient relies on differences in the abundance of a substance on the outside versus the inside of a cell and flows from areas of high to low ion concentration. In contrast, the electrical gradient revolves around an...
126.4K

También podría leer

Artículos Relacionados

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

Ordenar por
Same author

A light-initiated chemical reporter strategy for spatiotemporal labeling of biomolecules.

RSC chemical biology·2022
Same author

Downregulation of HULC Induces Ferroptosis in Hepatocellular Carcinoma via Targeting of the miR-3200-5p/ATF4 Axis.

Oxidative medicine and cellular longevity·2022
Same author

Effect of Programmed Nursing Plan Based on Thinking Map Guidance Mode on Hemodynamics and Intestinal Function Recovery of Patients Undergoing Endoscopic Retrograde Cholangiopancreatography.

Emergency medicine international·2022
Same author

In vivo non-invasive confocal fluorescence imaging beyond 1,700 nm using superconducting nanowire single-photon detectors.

Nature nanotechnology·2022
Same author

Primary signet ring cell carcinoma of the appendix: An interesting case.

The American journal of the medical sciences·2022
Same author

MYC drives autophagy to adapt to stress in Penaeus vannamei.

Fish & shellfish immunology·2022

Video Experimental Relacionado

Updated: Jan 8, 2026

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

983

Exploración de las vulnerabilidades del aprendizaje federado: un análisis en profundidad de los ataques de inversión

Pengxin Guo, Runxi Wang, Shuang Zeng

    IEEE transactions on pattern analysis and machine intelligence
    |December 22, 2025
    PubMed
    Resumen

    Los ataques de inversión de gradientes (GIA) amenazan la privacidad en el aprendizaje federado (FL). Este estudio categoriza los métodos GIA, encontrando que los GIA basados en optimización son los más prácticos pero imperfectos, mientras que otros son poco prácticos. Se proponen defensas para mejorar la seguridad de FL.

    Palabras clave:
    ataques de inversión de gradientesaprendizaje federadoprivacidad de datosseguridad de aprendizaje automáticodefensas de privacidad

    Videos de Experimentos Relacionados

    Last Updated: Jan 8, 2026

    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

    983

    Área de la Ciencia:

    • Aprendizaje automático
    • Ciberseguridad
    • Privacidad de datos

    Sus antecedentes:

    • El aprendizaje federado (FL) permite el entrenamiento colaborativo de modelos sin compartir datos brutos, pero la información del gradiente aún puede filtrar datos privados.
    • Los ataques de inversión de gradientes (GIA) representan un riesgo significativo para la privacidad en FL, pero faltan evaluaciones experimentales exhaustivas.

    Objetivo del estudio:

    • Revisar y categorizar sistemáticamente los ataques de inversión de gradientes (GIA) existentes en el aprendizaje federado (FL).
    • Analizar y evaluar experimentalmente la efectividad, practicidad y limitaciones de los diferentes tipos de GIA.
    • Proponer estrategias de defensa y direcciones de investigación futuras para una protección robusta de la privacidad en FL.

    Principales métodos:

    • Revisión sistemática de la literatura sobre métodos GIA.
    • Categorización de GIA en basados en optimización (OP-GIA), basados en generación (GEN-GIA) y basados en análisis (ANA-GIA).
    • Evaluación experimental exhaustiva de los tipos de GIA dentro de los marcos de FL.

    Principales resultados:

    • OP-GIA es el entorno de ataque más práctico, aunque su rendimiento es insatisfactorio.
    • GEN-GIA y ANA-GIA son generalmente poco prácticos debido a las dependencias y la detectabilidad, respectivamente.
    • Se identificaron factores clave que influyen en el rendimiento, la practicidad y los niveles de amenaza de GIA.

    Conclusiones:

    • Los métodos GIA existentes presentan grados variables de amenaza práctica para la privacidad de FL.
    • Se propone un pipeline de defensa de tres etapas para mejorar la seguridad del marco FL.
    • La investigación futura debe centrarse en el desarrollo de defensas más sólidas contra GIA sofisticados y en la exploración de nuevos vectores de ataque.