Domain Adaptation Learning with Kernel Logistic Regression and Kernel Canonical Correlation Analysis

LIU Jian-wei; SUN Zheng-kang; LIU Ze-yu; LUO Xiong-lin

doi:10.3969/j.issn.0372-2112.2016.12.014

您当前的位置：

首页 >

文章列表页 >

Domain Adaptation Learning with Kernel Logistic Regression and Kernel Canonical Correlation Analysis

更新时间：2025-07-16

- Domain Adaptation Learning with Kernel Logistic Regression and Kernel Canonical Correlation Analysis
- Acta Electronica Sinica Vol. 44, Issue 12, Pages: 2908-2915(2016)
- 作者机构：
  
  1. 中国石油大学(北京)自动化系,北京,102249
  2. 中国科学院软件研究所基础软件国家工程研究中心,北京,100190
  3. 中国石油大学(北京)自动化系,北京,102249
  4. 中国科学院软件研究所基础软件国家工程研究中心,北京,100190
- 作者简介：
- 基金信息：
- DOI：10.3969/j.issn.0372-2112.2016.12.014
  CLC： TP181
- Published Online：25 December 2016，
  
  Published：2016
- 稿件说明：
移动端阅览
LIU Jian-wei, SUN Zheng-kang, LIU Ze-yu, et al. Domain Adaptation Learning with Kernel Logistic Regression and Kernel Canonical Correlation Analysis[J]. Acta Electronica Sinica, 2016, 44(12): 2908-2915.
DOI：

LIU Jian-wei, SUN Zheng-kang, LIU Ze-yu, et al. Domain Adaptation Learning with Kernel Logistic Regression and Kernel Canonical Correlation Analysis[J]. Acta Electronica Sinica, 2016, 44(12): 2908-2915. DOI： 10.3969/j.issn.0372-2112.2016.12.014.

摘要

本文提出了一种利用核典型关联性分析提取源域目标域最大相关特征，使用核逻辑斯蒂回归模型进行域自适应学习的算法，该算法称为KCCA-DAML（Kernel Canonical Correlation Analysis for Domain Adaptation Learning）.该算法基于特征集关联性分析，有效的减小源域和目标域的概率分布差异性，利用提取的最大相关特征通过核逻辑斯蒂回归模型实现源域到目标域的跨域学习.实验比较源域数据上核逻辑斯蒂学习模型、目标域上核逻辑斯蒂学习模型、源域和目标域上核逻辑斯蒂学习模型和KCCA-DAML模型，结果显示KCCA-DAML在真实数据集上成功的实现了跨域学习.

Abstract

The domain adaptive learning algorithm using kernel logistic regression model is proposed.The proposed approach use kernel canonical correlation analysis to extract the maximum relevant features of the source and target domain.We dub it as KCCA-DAML(Kernel Canonical Correlation Analysis for Domain Adaptation Learning

KCCA-DAML).Our algorithm is based on canonical correlation analysis

which simultaneously minimizes the incompatibility among source features

target features and instance labels

extract maximum relevant features from source features

target features and instance labels

and use kernel logistic regression domain adaptation learning.In experimental comparison of the kernel logistic model and KCCA-DAML model on source domain data

the target domain data

source and the target domain data

we demonstrate the power of our techniques with the following real-world data sets:Reuters 20 Newsgroups

MNIST handwritten-digits and UCI Dermatology.

关键词

Keywords

references

Views

343

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Signal Sorting Method for Jagger PRI Radar Based on First-Order Difference Curve of DTOA and Correlation Analysis

Invariant Features Based Ship Detection Model for Multi-source Remote Sensing Images

Rejecting Crosstalk of Real-Time Ultrasonic Ranging System Based on Chaotic Pulse Position Modulation

Analysis and Design of Nonlinear Combining Functions in Stream Ciphers

Related Author

LIU Cheng-cheng

HU De-xiu

ZHAO Chuang

XIE Min

ZHANG Xin

GUO Hao-yuan

WANG Nan-nan

GAO Xin-bo

Related Institution

School of Data and Target Engineering，PLA Strategic Support Force Information Engineering University

State Key Laboratory of Integrated Services Networks, Xidian University

Chongqing Key Laboratory of Image Cognition, Chongqing University of Posts and Telecommunications

Tianjin Key Laboratory of Process Measurement and Control,School of Electrical Engineering and Automation,Tianjin University

Institute of Information Security,Xidian University

⁰