Abstract
Although neural models have performed impressively well on various tasks such as image recognition and question answering, their reasoning ability has been measured in only few studies. In this work, we focus on spatial reasoning and explore the spatial understanding of neural models. First, we describe the following two spatial reasoning IQ tests: rotation and shape composition. Using well-defined rules, we constructed datasets that consist of various complexity levels. We designed a variety of experiments in terms of generalization, and evaluated six different baseline models on the newly generated datasets. We provide an analysis of the results and factors that affect the generalization abilities of models. Also, we analyze how neural models solve spatial reasoning tests with visual aids. We hope that our work can encourage further research into human-level spatial reasoning and provide a new direction for future work.
Original language | English |
---|---|
Pages (from-to) | 27-38 |
Number of pages | 12 |
Journal | Neural Networks |
Volume | 140 |
DOIs | |
Publication status | Published - 2021 Aug |
Keywords
- Human IQ test
- Neural networks
- Spatial reasoning
ASJC Scopus subject areas
- Cognitive Neuroscience
- Artificial Intelligence