CRIENet / CrossRIEnet
====================

CRIENet implements the CrossRIEnet architecture described in:

Efstratios Manolakis, Christian Bongiorno, and Rosario N. Mantegna,
"Physics-Informed Singular-Value Learning for Cross-Covariances Forecasting
in Financial Markets", arXiv:2601.07687v3 (2026).

Relationship to RIEnet
----------------------

CrossRIEnet is a rectangular, singular-value-based companion to RIEnet's
square, eigenvalue-based rotationally invariant architecture.  The public API
and package organization intentionally follow RIEnet conventions where those
conventions are suitable for rectangular cross-correlation cleaning.

CRIENet is an independent package and does not import RIEnet or rely on RIEnet
private symbols.  Similar trainable building blocks reflect the shared
methodological lineage.  Git history does not establish that the current code
was copied from a particular RIEnet revision, so no stronger provenance claim
is made here.
