Semi-supervised Learning from General Unlabeled Data

Kaizhu Huang; Zenglin Xu; Irwin King; Michael R. Lyu

doi:10.1109/ICDM.2008.61

2008 Eighth IEEE International Conference on Data Mining

Semi-supervised Learning from General Unlabeled Data

Year: 2008, Pages: 273-282

DOI Bookmark: 10.1109/ICDM.2008.61

Authors

Kaizhu Huang
Zenglin Xu
Irwin King
Michael R. Lyu

Abstract

We consider the problem of Semi-supervised Learning (SSL) from general unlabeled data, which may contain irrelevant samples. Within the binary setting, our model manages to better utilize the information from unlabeled data by formulating them as a three-class (

- 1, + 1, 0

$-1,+1, 0$ ) mixture, where class

0

$0$ represents the irrelevant data. This distinguishes our work from the traditional SSL problem where unlabeled data are assumed to contain relevant samples only, either

+ 1

$+1$ or

- 1

$-1$ , which are forced to be the same

Like what you’re reading?

Already a member?

Get this article FREE with a new membership!

Learning from Positive and Unlabeled Examples: A Survey
2008 International Symposiums on Information Processing
Using Interacting Forces to Perform Semi-supervised Learning
Neural Networks, Brazilian Symposium on
Tri-Training Based Learning from Positive and Unlabeled Data
2008 International Symposiums on Information Processing
Semi-supervised Learning for Opinion Detection
Web Intelligence and Intelligent Agent Technology, IEEE/WIC/ACM International Conference on
Robustness Analysis of Network-Based Semi-supervised Learning Algorithms
Neural Networks, Brazilian Symposium on
Active Learning from Positive and Unlabeled Data
2011 IEEE 11th International Conference on Data Mining Workshops
SemiBoost: Boosting for Semi-Supervised Learning
IEEE Transactions on Pattern Analysis & Machine Intelligence
AuxMix: Semi-Supervised Learning with Unconstrained Unlabeled Data
2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
Unlabeled Backdoor Poisoning in Semi-Supervised Learning
2022 IEEE International Conference on Multimedia and Expo (ICME)
The Perils of Learning From Unlabeled Data: Backdoor Attacks on Semi-supervised Learning
2023 IEEE/CVF International Conference on Computer Vision (ICCV)

Semi-supervised Learning from General Unlabeled Data

Authors

Abstract

Related Articles