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

State of the art Deep Neural Networks (DNN) can now achieve above human levelaccuracy on image classification tasks. However their outstanding performancescome along with a complex inference mechanism making them arduouslyinterpretable models. In order to understand the underlying prediction rules ofDNNs, Dhamdhere et al. propose an interpretability method to break down a DNNprediction score as sum of its hidden unit contributions, in the form of ametric called conductance. Analyzing conductances of DNN hidden units, we findout there is a difference in how wrong and correct predictions are inferred. Weidentify distinguishable patterns of hidden unit activations for wrong andcorrect predictions. We then use an error detector in the form of a binaryclassifier on top of the DNN to automatically discriminate wrong and correctpredictions of the DNN based on their hidden unit activations. Detected wrongpredictions are discarded, increasing the model robustness. A differentapproach to distinguish wrong and correct predictions of DNNs is proposed byWang et al. whose method is based on the premise that input samples leading aDNN into making wrong predictions are less stable to the DNN weight changesthan correctly classified input samples. In our study, we compare both methodsand find out by combining them that better detection of wrong predictions canbe achieved.

Leveraging Model Interpretability and Stability to increase Model Robustness