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

For an explanation of a deep learning model to be effective, it must provideboth insight into a model and suggest a corresponding action in order toachieve some objective. Too often, the litany of proposed explainable deeplearning methods stop at the first step, providing practitioners with insightinto a model, but no way to act on it. In this paper, we propose contextualdecomposition explanation penalization (CDEP), a method which enablespractitioners to leverage existing explanation methods in order to increase thepredictive accuracy of deep learning models. In particular, when shown that amodel has incorrectly assigned importance to some features, CDEP enablespractitioners to correct these errors by directly regularizing the providedexplanations. Using explanations provided by contextual decomposition (CD)(Murdoch et al., 2018), we demonstrate the ability of our method to increaseperformance on an array of toy and real datasets.

Interpretations are useful: penalizing explanations to align neural networks with prior knowledge