English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

As convolutional neural networks (CNNs) become more accurate at objectrecognition, their representations become more similar to the primate visualsystem. This finding has inspired us and other researchers to ask if theimplication also runs the other way: If CNN representations become morebrain-like, does the network become more accurate? Previous attempts to addressthis question showed very modest gains in accuracy, owing in part tolimitations of the regularization method. To overcome these limitations, wedeveloped a new neural data regularizer for CNNs that uses Deep CanonicalCorrelation Analysis (DCCA) to optimize the resemblance of the CNN's imagerepresentations to that of the monkey visual cortex. Using this new neural dataregularizer, we see much larger performance gains in both classificationaccuracy and within-super-class accuracy, as compared to the previousstate-of-the-art neural data regularizers. These networks are also more robustto adversarial attacks than their unregularized counterparts. Together, theseresults confirm that neural data regularization can push CNN performancehigher, and introduces a new method that obtains a larger performance boost.

Improving the Accuracy and Robustness of CNNs Using a Deep CCA Neural Data Regularizer