Back to Search
Start Over
Multi-EPL: Accurate multi-source domain adaptation
- Source :
- PLoS ONE, Vol 16, Iss 8, p e0255754 (2021), PLoS ONE, Vol 16, Iss 8 (2021), PLoS ONE
- Publication Year :
- 2021
- Publisher :
- Public Library of Science (PLoS), 2021.
-
Abstract
- Given multiple source datasets with labels, how can we train a target model with no labeled data?Multi-source domain adaptation (MSDA) aims to train a model using multiple source datasets different from a target dataset in the absence of target data labels. MSDA is a crucial problem applicable to many practical cases where labels for the target data are unavailable due to privacy issues. Existing MSDA frameworks are limited since they align data without considering labels of the features of each domain. They also do not fully utilize the target data without labels and rely on limited feature extraction with a single extractor. In this paper, we proposeMulti-EPL, a novel method for MSDA.Multi-EPLexploits label-wise moment matching to align the conditional distributions of the features for the labels, uses pseudolabels for the unavailable target labels, and introduces an ensemble of multiple feature extractors for accurate domain adaptation. Extensive experiments show thatMulti-EPLprovides the state-of-the-art performance for MSDA tasks in both image domains and text domains, improving the accuracy by up to 13.20%.
- Subjects :
- Computer and Information Sciences
Neural Networks
Structural Engineering
Matching (graph theory)
Computer science
Science
Feature extraction
Datasets as Topic
Research and Analysis Methods
Domain (software engineering)
Image (mathematics)
Deep Learning
Electronics Engineering
Prototypes
Data Management
Multidisciplinary
Artificial neural network
business.industry
Applied Mathematics
Simulation and Modeling
Biology and Life Sciences
Pattern recognition
Conditional probability distribution
Built Structures
Technology Development
Feature (computer vision)
Physical Sciences
Database Management Systems
Engineering and Technology
Medicine
Artificial intelligence
Electronics
business
Mathematics
Algorithms
Network Analysis
Multi-source
Research Article
Neuroscience
Subjects
Details
- ISSN :
- 19326203
- Volume :
- 16
- Database :
- OpenAIRE
- Journal :
- PLOS ONE
- Accession number :
- edsair.doi.dedup.....0f0a54c9953eae543c91bb21354de2be