Controversial stimuli: Pitting neural networks against each other as models of human cognition
Open accessThe DOI matches its Crossref record; 28 of the 28 references it cites were found in a registry or an index.
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 ground-truth 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.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805
Image quality assessment: from error visibility to structural similarity
Wang, Z., Bovik, A. C., Sheikh, H. R., & Simoncelli, E. P. (2004) · IEEE Transactions on Image Processing · doi:10.1109/tip.2003.819861 · Cited by 57,892
The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
Zhang, R., Isola, P., Efros, A. A., Shechtman, E., & Wang, O. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00068 · Cited by 14,623
Explaining and Harnessing Adversarial Examples
Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014) · arXiv · doi:10.48550/arxiv.1412.6572 · Cited by 10,381
Intriguing properties of neural networks
Szegedy, C., Zaremba, W., Sutskever, I., Bruna, J., Erhan, D., Goodfellow, I., & Fergus, R. (2013) · arXiv · doi:10.48550/arxiv.1312.6199 · Cited by 8,143
Zagoruyko, S., & Komodakis, N. (2016) · Procedings of the British Machine Vision Conference 2016 · doi:10.5244/c.30.87 · Cited by 7,244
On calibration of modern neural networks
Guo, C., Pleiss, G., Sun, Y., & Weinberger, K. Q. (2017) · arXiv · doi:10.48550/arxiv.1706.04599 · Cited by 3,435
Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images
Nguyen, A., Yosinski, J., & Clune, J. (2014) · arXiv · doi:10.48550/arxiv.1412.1897 · Cited by 3,258
Deep neural networks are easily fooled: High confidence predictions for unrecognizable images
Nguyen, A., Yosinski, J., & Clune, J. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298640 · Cited by 3,258
Shortcut learning in deep neural networks
Geirhos, R., Jacobsen, J.-H., Michaelis, C., Zemel, R., Brendel, W., Bethge, M., & Wichmann, F. A. (2020) · Nature Machine Intelligence · doi:10.1038/s42256-020-00257-z · Cited by 2,224
Obfuscated gradients give a false sense of security: Circumventing defenses to adversarial examples
Athalye, A., Carlini, N., & Wagner, D. (2018) · arXiv · doi:10.48550/arxiv.1802.00420 · Cited by 2,172
Performance-optimized hierarchical models predict neural responses in higher visual cortex
Yamins, D. L. K., Hong, H., Cadieu, C. F., Solomon, E. A., Seibert, D., & DiCarlo, J. J. (2014) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1403112111 · Cited by 1,924
xarray: N-D labeled Arrays and Datasets in Python
Hoyer, S., & Hamman, J. (2017) · Journal of Open Research Software · doi:10.5334/jors.148 · Cited by 1,856
Using goal-driven deep learning models to understand sensory cortex
Yamins, D. L. K., & DiCarlo, J. J. (2016) · Nature Neuroscience · doi:10.1038/nn.4244 · Cited by 1,621
On a measure of the information provided by an experiment
Lindley, D. V. (1956) · The Annals of Mathematical Statistics · doi:10.1214/aoms/1177728069 · Cited by 1,509
Geirhos, R., Rubisch, P., Michaelis, C., Bethge, M., Wichmann, F. A., & Brendel, W. (2018) · arXiv · doi:10.48550/arxiv.1811.12231 · Cited by 1,490 · cited 1×
“, training a DNN on natural images and testing on silhouettes (7, 8)], as well as images degraded by noise or distortions (9–11), filtered (4), retextured (12), or adversarially perturbed to bias a DNN’s classifications (13).”
BackgroundBenchmarking Neural Network Robustness to Common Corruptions and Perturbations
Hendrycks, D., & Dietterich, T. (2019) · arXiv · doi:10.48550/arxiv.1903.12261 · Cited by 1,335
Delving into Transferable Adversarial Examples and Black-box Attacks
Liu, Y., Chen, X., Liu, C., & Song, D. (2016) · arXiv · doi:10.48550/arxiv.1611.02770 · Cited by 1,310
Deep supervised, but not unsupervised, models may explain IT cortical representation
Khaligh-Razavi, S.-M., & Kriegeskorte, N. (2014) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1003915 · Cited by 1,204
Self-citationKriegeskorte, N. (2015) · Annual Review of Vision Science · doi:10.1146/annurev-vision-082114-035447 · Cited by 1,093
Self-citationGuclu, U., & van Gerven, M. A. J. (2015) · Journal of Neuroscience · doi:10.1523/jneurosci.5023-14.2015 · Cited by 961
Cadieu, C. F., Hong, H., Yamins, D. L. K., Pinto, N., Ardila, D., Solomon, E. A., Majaj, N. J., & DiCarlo, J. J. (2014) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1003963 · Cited by 824
Cichy, R. M., Khosla, A., Pantazis, D., Torralba, A., & Oliva, A. (2016) · Scientific Reports · doi:10.1038/srep27755 · Cited by 716
Bayesian Active Learning for Classification and Preference Learning
Houlsby, N., Huszár, F., Ghahramani, Z., & Lengyel, M. (2011) · arXiv · doi:10.48550/arxiv.1112.5745 · Cited by 507 · cited 1×
“Ideally, we would take an optimal experimental-design approach (21, 22) and estimate, for a given image, how much seeing the response would reduce our uncertainty about which model generated the data (assuming that one of the models underlies the observed human responses).”
MethodSeeing it all: Convolutional network layers map the function of the human visual system
Eickenberg, M., Gramfort, A., Varoquaux, G., & Thirion, B. (2017) · NeuroImage · doi:10.1016/j.neuroimage.2016.10.001 · Cited by 338
Deep neural networks in computational neuroscience
Kietzmann, T. C., McClure, P., & Kriegeskorte, N. (2019) · Oxford Research Encyclopedia of Neuroscience · doi:10.1093/acrefore/9780190264086.013.46 · Cited by 186
Self-citationWang, Z., & Simoncelli, E. P. (2008) · Journal of Vision · doi:10.1167/8.12.8 · Cited by 138 · cited 1×
“Our approach is conceptually related to maximum differentiation (MAD) competition (18).”
BackgroundIndividual differences among deep neural network models
Mehrer, J., Spoerer, C. J., Kriegeskorte, N., & Kietzmann, T. C. (2020) · Nature Communications · doi:10.1038/s41467-020-19632-w · Cited by 133 · cited 1×
“An alternative approach to evaluating models would be to use multiple instances for each model (44), considering each DNN instance as an equivalent of an individual human brain.”
MethodSelf-citation
Cited by
Convolutional Neural Networks as a Model of the Visual System: Past, Present, and Future
Lindsay, G. W. (2021) · Journal of Cognitive Neuroscience · doi:10.1162/jocn_a_01544 · Cited by 557
The neural architecture of language: Integrative modeling converges on predictive processing
Schrimpf, M., Blank, I. A., Tuckute, G., Kauf, C., Hosseini, E. A., Kanwisher, N., Tenenbaum, J. B., & Fedorenko, E. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2105646118 · Cited by 517
Unsupervised neural network models of the ventral visual stream
Zhuang, C., Yan, S., Nayebi, A., Schrimpf, M., Frank, M. C., DiCarlo, J. J., & Yamins, D. L. K. (2021) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2014196118 · Cited by 337
Integrative benchmarking to advance neurally mechanistic models of human intelligence
Schrimpf, M., Kubilius, J., Lee, M. J., Ratan Murty, N. A., Ajemian, R., & DiCarlo, J. J. (2020) · Neuron · doi:10.1016/j.neuron.2020.07.040 · Cited by 233
The neuroconnectionist research programme
Doerig, A., Sommers, R. P., Seeliger, K., Richards, B., Ismael, J., Lindsay, G. W., Kording, K. P., Konkle, T., van Gerven, M. A. J., Kriegeskorte, N., et al. (2023) · Nature Reviews Neuroscience · doi:10.1038/s41583-023-00705-w · Cited by 233
Self-citation
Metadata
- Authors
- Golan, T., Raju, P. C., & Kriegeskorte, N.
- Venue
- Proceedings of the National Academy of Sciences
- Published
- 2020-11-23
- Volume
- 117
- Issue
- 47
- Pages
- 29330-29337
- ISSN
- 0027-8424 (ISSN-L)
- 1091-6490
- OpenAlex
- W3108490213
- Cited by
- 106
- References
- 28
- FWCI
- 6.23
- Type
- journal article
- Language
- en
- Open access
- green
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.