September 2019
Volume 19, Issue 10
Open Access
Vision Sciences Society Annual Meeting Abstract  |   September 2019
Learning to generalize like humans using basic-level object labels
Author Affiliations & Notes
  • Joshua C Peterson
    Department of Psychology, University of California, Berkeley
  • Paul Soulos
    Department of Psychology, University of California, Berkeley
  • Aida Nematzadeh
    Department of Psychology, University of California, Berkeley
  • Thomas L Griffiths
    Department of Psychology, University of California, Berkeley
Journal of Vision September 2019, Vol.19, 60a. doi:https://doi.org/10.1167/19.10.60a
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Joshua C Peterson, Paul Soulos, Aida Nematzadeh, Thomas L Griffiths; Learning to generalize like humans using basic-level object labels. Journal of Vision 2019;19(10):60a. https://doi.org/10.1167/19.10.60a.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Convolutional neural networks (CNNs) have become a standard for modeling several aspects of human visual processing, especially natural object classification, where they rival humans in performance. Most recent work on improving the correspondence between CNNs and humans has focused on low-level architectural modifications, and has paid less attention to changes in training supervision. We identify one way in which the training objective of the network differs greatly from that of humans: CNNs are almost exclusively trained on fine-grained, subordinate-level labels (e.g., Dalmatian), while humans also make use of more coarse-grained, basic-level labels (e.g., dog) that unify otherwise perceptually divergent subordinate classes. We show through a series of experiments that the level of abstraction in the labels used to train the network determines to a large extent how it will generalize, and consequently its correspondence with human generalization behavior.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×