Unsupervised Abstract Reasoning for Raven's Problem Matrices
arXiv:2109.10011
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.
arXiv commentAccepted by TIP
References in corpus (11)
- Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting
- Learning deep representations by mutual information estimation and maximization
- A Theoretical Analysis of Contrastive Unsupervised Representation Learning
- Measuring abstract reasoning in neural networks
- What Can Neural Networks Reason About?
- Learning from Complementary Labels
- Learning Perceptual Inference by Contrasting
- Abstract Reasoning with Distracting Features
- IQ of Neural Networks
- Solving Raven's Progressive Matrices with Neural Networks
- Stratified Rule-Aware Network for Abstract Visual Reasoning