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

Conventional frequentist learning is known to yield poorly calibrated modelsthat fail to reliably quantify the uncertainty of their decisions. Bayesianlearning can improve calibration, but formal guarantees apply only underrestrictive assumptions about correct model specification. Conformal prediction(CP) offers a general framework for the design of set predictors withcalibration guarantees that hold regardless of the underlying data generationmechanism. However, when training data are limited, CP tends to produce large,and hence uninformative, predicted sets. This paper introduces a novelmeta-learning solution that aims at reducing the set prediction size. Unlikeprior work, the proposed meta-learning scheme, referred to as meta-XB, (i)builds on cross-validation-based CP, rather than the less efficientvalidation-based CP; and (ii) preserves formal per-task calibration guarantees,rather than less stringent task-marginal guarantees. Finally, meta-XB isextended to adaptive non-conformal scores, which are shown empirically tofurther enhance marginal per-input calibration.

Few-Shot Calibration of Set Predictors via Meta-Learned Cross-Validation-Based Conformal Prediction