Back to Search
Start Over
Unsupervised Abstract Reasoning for Raven's Problem Matrices
- Source :
- IEEE transactions on image processing : a publication of the IEEE Signal Processing Society. 30
- Publication Year :
- 2021
-
Abstract
- Raven’s Progressive Matrices (RPM) is highly correlated with human intelligence, and it has been widely used to measure the abstract reasoning ability of humans. In this paper, to study the abstract reasoning capability of deep neural networks, we propose the first unsupervised learning method for solving RPM problems. Since the ground truth labels are not allowed, we design a pseudo target based on the prior constraints of the RPM formulation to approximate the ground-truth label, which effectively converts the unsupervised learning strategy into a supervised one. However, the correct answer is wrongly labelled by the pseudo target, and thus the noisy contrast will lead to inaccurate model training. To alleviate this issue, we propose to improve the model performance with negative answers. Moreover, we develop a decentralization method to adapt the feature representation to different RPM problems. Extensive experiments on three datasets demonstrate that our method even outperforms some of the supervised approaches. Our code is available at https://github.com/visiontao/ncd .
- Subjects :
- Intelligence Tests
Ground truth
Human intelligence
business.industry
Computer science
Supervised learning
Intelligence
Machine learning
computer.software_genre
Computer Graphics and Computer-Aided Design
Data modeling
Raven's Progressive Matrices
Feature (machine learning)
Unsupervised learning
Humans
Artificial intelligence
Neural Networks, Computer
Representation (mathematics)
business
computer
Software
Algorithms
Problem Solving
Subjects
Details
- ISSN :
- 19410042
- Volume :
- 30
- Database :
- OpenAIRE
- Journal :
- IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
- Accession number :
- edsair.doi.dedup.....da5d568862571928ec43a89372889afd