Adapting Deep Network Features to Capture Psychological Representations: An Abridged Report

Adapting Deep Network Features to Capture Psychological Representations: An Abridged Report

Joshua C. Peterson, Joshua T. Abbott, Thomas L. Griffiths

Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence
Best Sister Conferences. Pages 4934-4938. https://doi.org/10.24963/ijcai.2017/697

Deep neural networks have become increasingly successful at solving classic perception problems (e.g., recognizing objects), often reaching or surpassing human-level accuracy. In this abridged report of Peterson et al. [2016], we examine the relationship between the image representations learned by these networks and those of humans. We find that deep features learned in service of object classification account for a significant amount of the variance in human similarity judgments for a set of animal images. However, these features do not appear to capture some key qualitative aspects of human representations. To close this gap, we present a method for adapting deep features to align with human similarity judgments, resulting in image representations that can potentially be used to extend the scope of psychological experiments and inform human-centric AI.
Keywords:
Artificial Intelligence: knowledge representation and reasoning
Artificial Intelligence: cognitive science
Artificial Intelligence: computer vision
Artificial Intelligence: artificial intelligence