Abstract
Same-different visual reasoning is a basic skill central to abstract combinatorial thought. This fact has lead neural networks researchers to test same-different classification on deep convolutional neural networks (DCNNs), which has resulted in a controversy regarding whether this skill is within the capacity of these models. However, most tests of same-different classification rely on testing on images that come from the same pixel-level distribution as the testing images, yielding the results inconclusive. In this study we tested relational same-different reasoning DCNNs. In a series of simulations we show that models based on the ResNet-50 architecture are capable of visual same-different classification, but only when the test images are similar to the training images at the pixel-level. In contrast, even when there are only subtle differences between the testing and training images, the performance of DCNNs drops substantially. This is true even when DCNNs’ training regime is augmented with images from new versions of the same-different task or through multi-task learning on the test images. Furthermore, we show that the Relation Network, a deep learning architecture specifically designed to tackle visual relational reasoning problems, suffers the same kind of limitations than ResNet-50 classifiers.
Competing Interest Statement
The authors have declared no competing interest.
