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

Binary classification is widely used in ML production systems. Monitoringclassifiers in a constrained event space is well known. However, real worldproduction systems often lack the ground truth these methods require. Privacyconcerns may also require that the ground truth needed to evaluate theclassifiers cannot be made available. In these autonomous settings,non-parametric estimators of performance are an attractive solution. They donot require theoretical models about how the classifiers made errors in anygiven sample. They just estimate how many errors there are in a sample of anindustrial or robotic datastream. We construct one such non-parametricestimator of the sample errors for an ensemble of weak binary classifiers. Ourapproach uses algebraic geometry to reformulate the self-assessment problem forensembles of binary classifiers as an exact polynomial system. The polynomialformulation can then be used to prove - as an algebraic geometry algorithm -that no general solution to the self-assessment problem is possible. However,specific solutions are possible in settings where the engineering context putsthe classifiers close to independent errors. The practical utility of themethod is illustrated on a real-world dataset from an online advertisingcampaign and a sample of common classification benchmarks. The accuracyestimators in the experiments where we have ground truth are better than onepart in a hundred. The online advertising campaign data, where we do not haveground truth data, is verified by an internal consistency approach whosevalidity we conjecture as an algebraic geometry theorem. We call this approach- algebraic ground truth inference.

Algebraic Ground Truth Inference: Non-Parametric Estimation of Sample Errors by AI Algorithms