ConceptFDR / False Discovery Rate
False Discovery Rate
In multiple hypothesis testing, the average proportion of selected variables that are actually noise (false positives). It is an important metric in frequentist statistics for controlling erroneous discoveries.
Timeline (last 90 days)
Oct 16
Research on SelectPFN proved a negative result: training in the average case does not automatically translate into frequentist guarantees under a fixed ground truth, so controlling the false discovery rate on any given dataset requires additional mechanisms.
Unverified50%