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

Multivariate techniques and machine learning models have found numerousapplications in High Energy Physics (HEP) research over many years. In recenttimes, AI models based on deep neural networks are becoming increasinglypopular for many of these applications. However, neural networks are regardedas black boxes -- because of their high degree of complexity it is often quitedifficult to quantitatively explain the output of a neural network byestablishing a tractable input-output relationship and information propagationthrough the deep network layers. As explainable AI (xAI) methods are becomingmore popular in recent years, we explore interpretability of AI models byexamining an Interaction Network (IN) model designed to identify boosted $H\tob\bar{b}$ jets amid QCD background. We explore different quantitative methodsto demonstrate how the classifier network makes its decision based on theinputs and how this information can be harnessed to reoptimize the model-makingit simpler yet equally effective. We additionally illustrate the activity ofhidden layers within the IN model as Neural Activation Pattern (NAP) diagrams.Our experiments suggest NAP diagrams reveal important information about howinformation is conveyed across the hidden layers of deep model. These insightscan be useful to effective model reoptimization and hyperparameter tuning.

Interpretability of an Interaction Network for identifying $H \rightarrow b\bar{b}$ jets