Physics-Informed Singular-Value Learning for Cross-Covariances Forecasting in Financial Markets

By Efstratios Manolakis, Christian Bongiorno, Rosario N. Mantegna

Published 2026-08-02

Everscope rating

2002.6

Relevance to quantitative trading

9 / 10

Implementation complexity

6 / 10

Reproducibility

4 / 5

About this paper

Methodology: Physics-Informed Neural Singular-Value Cleaning (Two-Stream Architecture). Problem types: Portfolio Optimization, Risk Management, Time Series Forecasting, Dimensionality Reduction, Optimization.

arXiv:2601.07687 · Code · Paper rankings

Open the interactive Everscope explorer for full analysis, charts, and paper battles.