Controversial stimuli: Pitting neural networks against each other as models of human cognition

Tal Golan, Prashant C. Raju, Nikolaus Kriegeskorte

Research output: Contribution to journalArticlepeer-review

47 Scopus citations

Abstract

Distinct scientific theories can make similar predictions. To adjudicate between theories, we must design experiments for which the theories make distinct predictions. Here we consider the problem of comparing deep neural networks as models of human visual recognition. To efficiently compare models' ability to predict human responses, we synthesize controversial stimuli: images for which different models produce distinct responses.We applied this approach to two visual recognition tasks, handwritten digits (MNIST) and objects in small natural images (CIFAR-10). For each task, we synthesized controversial stimuli to maximize the disagreement among models which employed different architectures and recognition algorithms. Human subjects viewed hundreds of these stimuli, as well as natural examples, and judged the probability of presence of each digit/object category in each image. We quantified how accurately each model predicted the human judgments. The best-performing models were a generative analysis-by-synthesis model (based on variational autoencoders) for MNIST and a hybrid discriminative-generative joint energy model for CIFAR-10. These deep neural networks (DNNs), which model the distribution of images, performed better than purely discriminative DNNs, which learn only to map images to labels. None of the candidate models fully explained the human responses. Controversial stimuli generalize the concept of adversarial examples, obviating the need to assume a groundtruth model. Unlike natural images, controversial stimuli are not constrained to the stimulus distribution models are trained on, thus providing severe out-of-distribution tests that reveal the models' inductive biases. Controversial stimuli therefore provide powerful probes of discrepancies between models and human perception.

Original languageEnglish
Pages (from-to)29330-29337
Number of pages8
JournalProceedings of the National Academy of Sciences of the United States of America
Volume117
Issue number47
DOIs
StatePublished - 24 Nov 2020
Externally publishedYes

Keywords

  • Adversarial examples
  • Deep neural networks
  • Generative modeling
  • Optimal experimental design
  • Visual object recognition

ASJC Scopus subject areas

  • General

Fingerprint

Dive into the research topics of 'Controversial stimuli: Pitting neural networks against each other as models of human cognition'. Together they form a unique fingerprint.

Cite this