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

Despite their successful application to a variety of tasks, neural networksremain limited, like other machine learning methods, by their sensitivity toshifts in the data: their performance can be severely impacted by differencesin distribution between the data on which they were trained and that on whichthey are deployed. In this article, we propose a new family of representations,called MAGDiff, that we extract from any given neural network classifier andthat allows for efficient covariate data shift detection without the need totrain a new model dedicated to this task. These representations are computed bycomparing the activation graphs of the neural network for samples belonging tothe training distribution and to the target distribution, and yield powerfuldata- and task-adapted statistics for the two-sample tests commonly used fordata set shift detection. We demonstrate this empirically by measuring thestatistical powers of two-sample Kolmogorov-Smirnov (KS) tests on severaldifferent data sets and shift types, and showing that our novel representationsinduce significant improvements over a state-of-the-art baseline relying on thenetwork output.

MAGDiff: Covariate Data Set Shift Detection via Activation Graphs of Deep Neural Networks