• Home
  • Search
  • Minimum Class Confusion for Versatile Domain Adaptation
  • Open Access IconOpen Access
  • Cite Icon372
  • https://doi.org/10.1007/978-3-030-58589-1_28Copy DOI Icon

Minimum Class Confusion for Versatile Domain Adaptation

  • Jan 1, 2020
  • Ying Jin +3 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

There are a variety of Domain Adaptation (DA) scenarios subject to label sets and domain configurations, including closed-set and partial-set DA, as well as multi-source and multi-target DA. It is notable that existing DA methods are generally designed only for a specific scenario, and may underperform for scenarios they are not tailored to. To this end, this paper studies Versatile Domain Adaptation (VDA), where one method can handle several different DA scenarios without any modification. Towards this goal, a more general inductive bias other than the domain alignment should be explored. We delve into a missing piece of existing methods: class confusion, the tendency that a classifier confuses the predictions between the correct and ambiguous classes for target examples, which is common in different DA scenarios. We uncover that reducing such pairwise class confusion leads to significant transfer gains. With this insight, we propose a general loss function: Minimum Class Confusion (MCC). It can be characterized as (1) a non-adversarial DA method without explicitly deploying domain alignment, enjoying faster convergence speed; (2) a versatile approach that can handle four existing scenarios: Closed-Set, Partial-Set, Multi-Source, and Multi-Target DA, outperforming the state-of-the-art methods in these scenarios, especially on one of the largest and hardest datasets to date ( $$7.3\%$$ on DomainNet). Its versatility is further justified by two scenarios proposed in this paper: Multi-Source Partial DA and Multi-Target Partial DA. In addition, it can also be used as a general regularizer that is orthogonal and complementary to a variety of existing DA methods, accelerating convergence and pushing these readily competitive methods to stronger ones. Code is available at https://github.com/thuml/Versatile-Domain-Adaptation .

Similar Papers
  • Research Article
  • Citations99

A Stacked Auto-Encoder Based Partial Adversarial Domain Adaptation Model for Intelligent Fault Diagnosis of Rotating Machines

  • Dec 15, 2020
  • IEEE Transactions on Industrial Informatics
  • Zhao-Hua Liu +5
  • Research Article
  • Citations50

Partial Transfer Fault Diagnosis by Multiscale Weight-Selection Adversarial Network

  • Dec 01, 2022
  • IEEE/ASME Transactions on Mechatronics
  • Quan Qian +3
  • Book Chapter
  • Citations541

Partial Adversarial Domain Adaptation

  • Jan 01, 2018
  • Zhangjie Cao +3
  • Research Article
  • Citations13

One Fits Many: Class Confusion Loss for Versatile Domain Adaptation.

  • Nov 01, 2024
  • IEEE transactions on pattern analysis and machine intelligence
  • Ying Jin +4
  • Research Article
  • Citations50

On statistic alignment for domain adaptation in structural health monitoring

  • Jul 08, 2022
  • Structural Health Monitoring
  • Jack Poole +4
  • Conference Article
  • Citations2

Dual Contrastive Universal Adaptation Network

  • Jul 05, 2021
  • Ziyun Cai +4
  • Research Article
  • Citations15

Manifold discrimination partial adversarial domain adaptation

  • Jun 25, 2022
  • Knowledge-Based Systems
  • Chunmei He +4
  • Research Article
  • Citations25

Multilevel Heterogeneous Domain Adaptation Method for Remote Sensing Image Segmentation

  • Jan 01, 2023
  • IEEE Transactions on Geoscience and Remote Sensing
  • Chenbin Liang +4
  • Research Article
  • Citations6

Partial domain adaptation based on shared class oriented adversarial network

  • Jun 20, 2020
  • Computer Vision and Image Understanding
  • Wenjie Qiu +2
  • Research Article
  • Citations24

Semi-supervised domain adaptation for segmentation models on different monitoring settings

  • May 01, 2023
  • Automation in Construction
  • Yeji Hong +4
  • Research Article
  • Citations31

PCLUDA: A Pseudo-Label Consistency Learning- Based Unsupervised Domain Adaptation Method for Cross-Domain Optical Remote Sensing Image Retrieval

  • Jan 01, 2023
  • IEEE Transactions on Geoscience and Remote Sensing
  • Dongyang Hou +3
  • Conference Article
  • Citations26

Partial Video Domain Adaptation with Partial Adversarial Temporal Attentive Network

  • Oct 01, 2021
  • Yuecong Xu +5
  • Research Article
  • Citations3

An iterative-cluster domain adaptation fault diagnosis method for incremental multi-target domain adaptation

  • Feb 25, 2025
  • Measurement Science and Technology
  • Cheng Wang +2
  • Research Article
  • Citations62

Does BERT need domain adaptation for clinical negation detection?

  • Feb 11, 2020
  • Journal of the American Medical Informatics Association
  • Chen Lin +5
  • Research Article
  • Citations18

A New Multisensor Partial Domain Adaptation Method for Machinery Fault Diagnosis Under Different Working Conditions

  • Jan 01, 2023
  • IEEE Transactions on Instrumentation and Measurement
  • Jun Zhu +4
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.