Human alignment of neural network representations
This work addresses the problem of understanding how neural networks align with human cognition for researchers in AI and cognitive science, but it is incremental as it builds on existing comparisons without introducing new methods.
The study investigated factors affecting alignment between neural network representations and human mental representations, finding that training dataset and objective function have a larger impact than model scale or architecture, with some concepts like food and animals well-represented but others like royal or sports-related objects not, and scaling alone is insufficient for matching human conceptual representations.
Today's computer vision models achieve human or near-human level performance across a wide variety of vision tasks. However, their architectures, data, and learning algorithms differ in numerous ways from those that give rise to human vision. In this paper, we investigate the factors that affect the alignment between the representations learned by neural networks and human mental representations inferred from behavioral responses. We find that model scale and architecture have essentially no effect on the alignment with human behavioral responses, whereas the training dataset and objective function both have a much larger impact. These findings are consistent across three datasets of human similarity judgments collected using two different tasks. Linear transformations of neural network representations learned from behavioral responses from one dataset substantially improve alignment with human similarity judgments on the other two datasets. In addition, we find that some human concepts such as food and animals are well-represented by neural networks whereas others such as royal or sports-related objects are not. Overall, although models trained on larger, more diverse datasets achieve better alignment with humans than models trained on ImageNet alone, our results indicate that scaling alone is unlikely to be sufficient to train neural networks with conceptual representations that match those used by humans.