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

Over the past decade, deep neural networks have proven to be adept in imageclassification tasks, often surpassing humans in terms of accuracy. However,standard neural networks often fail to understand the concept of hierarchicalstructures and dependencies among different classes for vision related tasks.Humans on the other hand, seem to intuitively learn categories conceptually,progressively growing from understanding high-level concepts down to granularlevels of categories. One of the issues arising from the inability of neuralnetworks to encode such dependencies within its learned structure is that ofsubpopulation shift -- where models are queried with novel unseen classes takenfrom a shifted population of the training set categories. Since the neuralnetwork treats each class as independent from all others, it struggles tocategorize shifting populations that are dependent at higher levels of thehierarchy. In this work, we study the aforementioned problems through the lensof a novel conditional supervised training framework. We tackle subpopulationshift by a structured learning procedure that incorporates hierarchicalinformation conditionally through labels. Furthermore, we introduce a notion ofgraphical distance to model the catastrophic effect of mispredictions. We showthat learning in this structured hierarchical manner results in networks thatare more robust against subpopulation shifts, with an improvement up to 3\% interms of accuracy and up to 11\% in terms of graphical distance over standardmodels on subpopulation shift benchmarks.

Encoding Hierarchical Information in Neural Networks helps in Subpopulation Shift