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

Machine Learning explainability techniques have been proposed as a means of`explaining' or interrogating a model in order to understand why a particulardecision or prediction has been made. Such an ability is especially importantat a time when machine learning is being used to automate decision processeswhich concern sensitive factors and legal outcomes. Indeed, it is even arequirement according to EU law. Furthermore, researchers concerned withimposing overly restrictive functional form (e.g., as would be the case in alinear regression) may be motivated to use machine learning algorithms inconjunction with explainability techniques, as part of exploratory research,with the goal of identifying important variables which are associated with anoutcome of interest. For example, epidemiologists might be interested inidentifying `risk factors' - i.e. factors which affect recovery from disease -by using random forests and assessing variable relevance using importancemeasures. However, and as we demonstrate, machine learning algorithms are notas flexible as they might seem, and are instead incredibly sensitive to theunderling causal structure in the data. The consequences of this are thatpredictors which are, in fact, critical to a causal system and highlycorrelated with the outcome, may nonetheless be deemed by explainabilitytechniques to be unrelated/unimportant/unpredictive of the outcome. Rather thanthis being a limitation of explainability techniques per se, we show that it israther a consequence of the mathematical implications of regression, and theinteraction of these implications with the associated conditionalindependencies of the underlying causal structure. We provide some alternativerecommendations for researchers wanting to explore the data for importantvariables.

Trying to Outrun Causality with Machine Learning: Limitations of Model Explainability Techniques for Identifying Predictive Variables