logo
科言猫学术研究的AI总结
language
手机端手机端
  • Phone Feature微信小程序
简体中文简体中文
notice消息通知
博客博客
arrow
返回

Federated discretization for supervised classification

delete2026-08-29
delete0
delete
OA
AI
J
José Luis Flores *
B
Borja Calvo
A
Aritz Pérez
DOI：10.1007/s10618-026-01247-2delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
几种分类器仅对离散变量进行操作，需要通过离散化预处理连续属性。监督离散化通过将连续变量的定义域划分为定义离散化策略的区间来满足这一需求。联邦学习则在不共享原始数据的情况下实现全局模型的协同训练：用户训练本地模型并将参数传递给中央服务器，服务器将其聚合为全局模型。在此设置下的关键挑战是在最小化通信开销的同时实现与集中式方法相当的性能。本文提出了一种联邦离散化框架，可在单次通信回合中计算全局离散化策略。每个用户为连续变量估计高斯混合模型并传输其参数（称为代表值）至服务器。服务器随后应用离散化算法对这些代表值进行操作以推导全局策略，随后将其分发给用户。所提出的框架可容纳多种离散化策略；具体而言，我们提出了等宽、等频、基于熵和基于核密度的方法的联邦适应性。在十个基准数据集上进行的实验表明，所提出的框架实现了与集中式离散化方法相当的性能，其中基于核密度的和等频变体获得了最具竞争力的结果。
Keyword:
Discretization
Supervised
Federated
Kernel density
AI总结

AI总结

对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

Data Mining and Knowledge Discovery 封面图
Data Mining and Knowledge Discovery
IF:
4.3
论文数:
224
被引数:
6.0K

机构

B
basque center of applied mathematics
学者数:
2
论文数: 1
被引数: 0
D
department of computer science and artificial intelligence
学者数:
61
论文数: 29
被引数: 0
引用论文

引用论文

Tiny Federated Learning with Bayesian Classifiers基于贝叶斯分类器的微型联邦学习
err2023 IEEE 32nd International Symposium on Industrial Electronics (ISIE)
IF0
err2023-06-19
err0
PREAI
errNing Xiong; Sasikumar Punnekkat
err分享
err收藏
err
IF0
err
err0
errOAAI
err
err分享
err收藏
Discrete Bayesian Network Classifiers: A Survey
errACM COMPUTING SURVEYS
IF28
err2014-07-14
err198
errOAAI
errBielza, Concha; Larranaga, Pedro
err分享
err收藏
DISCOLEAF: Personalized DIScretization of COntinuous Attributes for LEArning with Federated Decision Trees
errLecture Notes in Computer Science
IF0
err2024-01-01
err0
PREAI
errKwatra,Saloni; Torra,Vicenç
err分享
err收藏
A Survey on Deep Learning Techniques for Privacy-Preserving
errLecture Notes in Computer Science
IF0
err2019-09-09
err0
PREAI
errHarry Chandra Tanuwidjaja; Rakyong Choi; Kwangjo Kim
err分享
err收藏
err
IF0
err
err0
errOAAI
err
err分享
err收藏
A comparison study of centralized and decentralized federated learning approaches utilizing the transformer architecture for estimating remaining useful life
errRELIABILITY ENGINEERING & SYSTEM SAFETY
IF11
err2023-05-01
err29
PREAI
errKamei, Sayaka; Taghipour, Sharareh
err分享
err收藏
err
IF0
err
err0
errOAAI
err
err分享
err收藏
Human Activity Recognition Using Federated Learning
err2018 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Ubiquitous Computing & Communications, Big Data & Cloud Computing, Social Computing & Networking, Sustainable Computing & Communications (ISPA/IUCC/BDCloud/SocialCom/SustainCom)
IF0
err2018-12-01
err0
PREAI
errKonstantin Sozinov; Vladimir Vlassov; Sarunas Girdzijauskas
err分享
err收藏
Equal-bin-width histogram versus equal-bin-count histogram
errJournal of Applied Statistics
IF0
err2020-06-26
err0
errOAAI
errPiotr Sulewski
err分享
err收藏
学者 查看更多内容