Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Weighted Mean

Weighted Mean

While taking the arithmetic, geometric, or harmonic mean of a sample data set, equal importance is assigned to all the data points. However, all the values may not always be equally important in some data sets. An intrinsic bias might make it more important to give more weightage to specific values over others.
For example, consider the number of goals scored in the matches of a tournament. While computing the average number of goals scored in the tournament, it may be more important to...

What is Central Tendency?

What is Central Tendency?

Descriptive statistics describe or summarize relevant characteristics of a sample and aid in the analysis of data of interest. When analyzing large quantities of data and developing an inference, one needs to identify a value representative of the entire data set. Characteristics such as central tendency, extreme values, range of measurements, or the most repeated value can help better understand the data.
The central tendency is the most conventionally used data characteristic. It is a...

Cluster Sampling Method

Cluster Sampling Method

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...

Distributed Loads

Distributed Loads

Distributed loads are a common type of load that engineers and scientists encounter in various practical situations. Distributed loads often refer to a type of load spread over a surface or a structure and can be modeled as continuous force per unit area.
For example, consider a bookshelf filled with books stacked vertically adjacent to each other. The weight of the books is evenly distributed over the length of the shelf. As a result, the pressure at different locations on the surface of the...

Maximum Size of Aggregate

Maximum Size of Aggregate

The maximum size of aggregate is defined as the aperture of the sieve retaining 15 percent or more of the particles present in the aggregate sample. The aggregate's maximum size impacts the concrete's water requirement, workability, and strength. Larger aggregates reduce the surface area needing cement paste coverage, which can lower water needs, thereby allowing a decrease in the water-to-cement ratio when the desired workability and richness of the mix are to be maintained, which can...

Area Computation by the Alternative Coordinate Method

Area Computation by the Alternative Coordinate Method

The alternative coordinate method, also known as the Shoelace Formula, is a technique for determining the area of a traverse using Cartesian coordinates. This method relies on the sequential arrangement of x and y coordinates for each point of the shape, ensuring accuracy and ease of application.In this approach, each corner's x and y coordinates are listed as fractions, with the x-coordinate as the numerator and the y-coordinate as the denominator. These coordinates are arranged sequentially...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

In situ dynamic modulation of zero-valent and low-valent copper ratio for constructing stable copper catalysts for acetylene hydrochlorination.

Journal of colloid and interface science·2026

Same author

Stabilizing Lattice Oxygen Redox Through Bicarbonate Pyrolysis-Driven Multifunctional Interface Engineering in Li-Rich Layered Oxides.

Small (Weinheim an der Bergstrasse, Germany)·2026

Same author

Generation of magnonic frequency combs in a <i>PT</i>-symmetric cavity magnomechanical system.

Optics express·2026

Same author

Isolation, Identification, Biological Characteristics, and In Vitro and In Vivo Antibacterial Effects of a Bovine-Derived <i>Escherichia coli</i> Bacteriophage XJA18.

Microorganisms·2026

Same author

Predicting intracranial angioplasty failure using vessel wall MRI habitat radiomics and deep learning: a multicenter study.

Journal of neurointerventional surgery·2026

Same author

The Mechanisms and Research Progress of Electroacupuncture Modulation of the Anterior Cingulate Cortex in Alleviating Paclitaxel-induced Pain and Cognitive Impairment.

Current neuropharmacology·2026

Same journal

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026

Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026

See all related articles

Search research articles

Related Experiment Video

Updated: Sep 2, 2025

Averaging of Viral Envelope Glycoprotein Spikes from Electron Cryotomography Reconstructions using Jsubtomo

Averaging of Viral Envelope Glycoprotein Spikes from Electron Cryotomography Reconstructions using Jsubtomo

Published on: October 21, 2014

Decentralized Federated Averaging.

Tao Sun, Dongsheng Li, Bao Wang

IEEE Transactions on Pattern Analysis and Machine Intelligence

|August 4, 2022

Summary

This summary is machine-generated.

Decentralized federated averaging with momentum (DFedAvgM) enhances privacy and reduces communication costs in distributed machine learning. This novel approach outperforms traditional federated averaging (FedAvg) in convergence speed and efficiency.

More Related Videos

Cryo-Electron Tomography Remote Data Collection and Subtomogram Averaging

Cryo-Electron Tomography Remote Data Collection and Subtomogram Averaging

Published on: July 12, 2022

Analysis of SEC-SAXS data via EFA deconvolution and Scatter

Analysis of SEC-SAXS data via EFA deconvolution and Scatter

Published on: January 28, 2021

Related Experiment Videos

Last Updated: Sep 2, 2025

Averaging of Viral Envelope Glycoprotein Spikes from Electron Cryotomography Reconstructions using Jsubtomo

Averaging of Viral Envelope Glycoprotein Spikes from Electron Cryotomography Reconstructions using Jsubtomo

Published on: October 21, 2014

Cryo-Electron Tomography Remote Data Collection and Subtomogram Averaging

Cryo-Electron Tomography Remote Data Collection and Subtomogram Averaging

Published on: July 12, 2022

Analysis of SEC-SAXS data via EFA deconvolution and Scatter

Analysis of SEC-SAXS data via EFA deconvolution and Scatter

Published on: January 28, 2021

Area of Science:

Distributed machine learning
Optimization algorithms
Networked systems

Background:

Federated averaging (FedAvg) is a popular distributed training method, but its centralized nature poses communication bottlenecks and privacy risks.
Existing decentralized methods often lack the efficiency and robustness required for large-scale deployments.
The need for communication-efficient and privacy-preserving distributed learning algorithms is critical.

Purpose of the Study:

To introduce and analyze a decentralized federated averaging algorithm with momentum (DFedAvgM) for improved distributed training.
To investigate the convergence properties and communication efficiency of DFedAvgM, including its quantized variant.
To address the limitations of centralized federated learning by enabling peer-to-peer client communication.

Main Methods:

Implementation of decentralized federated averaging with momentum (DFedAvgM) on clients connected via an undirected graph.
Incorporation of multiple local iterations, a mixing matrix, and quantization to reduce communication overhead.
Lyapunov analysis to rigorously prove convergence under challenging conditions, extending previous theoretical frameworks.

Main Results:

Convergence of the proposed (quantized) DFedAvgM is proven under general assumptions.
Sublinear convergence rates are achieved when the loss function satisfies the PŁ property.
Numerical experiments demonstrate that DFedAvgM surpasses traditional FedAvg in both convergence speed and communication efficiency.

Conclusions:

DFedAvgM offers a more robust, private, and communication-efficient alternative to centralized federated averaging.
The algorithm's theoretical convergence guarantees and practical performance improvements highlight its potential for large-scale distributed learning.
Quantization further enhances communication efficiency without compromising convergence, making DFedAvgM suitable for resource-constrained environments.