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

Visual object recognition has been extensively studied in both neuroscienceand computer vision. Recently, the most popular class of artificial systems forthis task, deep convolutional neural networks (CNNs), has been shown to provideexcellent models for its functional analogue in the brain, the ventral streamin visual cortex. This has prompted questions on what, if any, are the commonprinciples underlying the reformatting of visual information as it flowsthrough a CNN or the ventral stream. Here we consider some prominentstatistical patterns that are known to exist in the internal representations ofeither CNNs or the visual cortex and look for them in the other system. We showthat intrinsic dimensionality (ID) of object representations along the rathomologue of the ventral stream presents two distinct expansion-contractionphases, as previously shown for CNNs. Conversely, in CNNs, we show thattraining results in both distillation and active pruning (mirroring theincrease in ID) of low- to middle-level image information in single units, asrepresentations gain the ability to support invariant discrimination, inagreement with previous observations in rat visual cortex. Taken together, ourfindings suggest that CNNs and visual cortex share a similarly tightrelationship between dimensionality expansion/reduction of objectrepresentations and reformatting of image information.

Prune and distill: similar reformatting of image information along rat visual cortex and deep neural networks