Nuisance Hardened Data Compression for Fast Likelihood-Free Inference

Alsing & Wandelt (2019) — nuisance-hardened compression

Result

projects the score into a locally/asymptotically Fisher-preserving summary for parameters of interest using the nuisance Schur geometry.

Does not transfer

the paper studies continuous summaries and does not prove Loewner domination for every hard quantizer or the project’s global $D_s$ upper bound.

Transfer

closest published representation-level antecedent of the full-data efficient score used by DS-EFFICIENT-SCORE-DOMINATION.