Abstract
Early in development, children learn to extend novel category labels to objects with the same shape, a phenomenon known as the shape bias. Inspired by these findings, Geirhos et al. (2019) examined whether deep neural networks show a shape or texture bias by constructing images with conflicting shape and texture cues. They found that convolutional neural networks strongly preferred to classify familiar objects based on texture as opposed to shape, suggesting a texture bias. However, there are a number of differences between how the networks were tested in this study versus how children are typically tested. In this work, we re-examine the inductive biases of neural networks by adapting the stimuli and procedure from Geirhos et al. (2019) to more closely follow the developmental paradigm and test on a wide range of pre-trained neural networks. Across three experiments, we find that deep neural networks exhibit a preference for shape rather than texture when tested under conditions that more closely replicate the developmental procedure.
Original language | English (US) |
---|---|
Pages | 1284-1290 |
Number of pages | 7 |
State | Published - 2022 |
Event | 44th Annual Meeting of the Cognitive Science Society: Cognitive Diversity, CogSci 2022 - Toronto, Canada Duration: Jul 27 2022 → Jul 30 2022 |
Conference
Conference | 44th Annual Meeting of the Cognitive Science Society: Cognitive Diversity, CogSci 2022 |
---|---|
Country/Territory | Canada |
City | Toronto |
Period | 7/27/22 → 7/30/22 |
Keywords
- inductive bias
- neural networks
- shape bias
- word learning
ASJC Scopus subject areas
- Artificial Intelligence
- Computer Science Applications
- Human-Computer Interaction
- Cognitive Neuroscience