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

Recurrent neural networks (RNNs) have been shown to perform better thanfeedforward architectures in visual object categorization tasks, especially inchallenging conditions such as cluttered images. However, little is known aboutthe exact computational role of recurrent information flow in these conditions.Here we test RNNs trained for object categorization on the hypothesis thatrecurrence iteratively aids object categorization via the communication ofcategory-orthogonal auxiliary variables (the location, orientation, and scaleof the object). Using diagnostic linear readouts, we find that: (a) informationabout auxiliary variables increases across time in all network layers, (b) thisinformation is indeed present in the recurrent information flow, and (c) itsmanipulation significantly affects task performance. These observations confirmthe hypothesis that category-orthogonal auxiliary variable information isconveyed through recurrent connectivity and is used to optimize categoryinference in cluttered environments.

Category-orthogonal object features guide information processing in recurrent neural networks trained for object categorization