STO-OPT-MATH-001: Generalized Identifiability, Masking Cones and Conditional Robustness in Phase–Memory Diagnostics of Optical Vortex Annihilation — Final Validated Mathematical Core v0.2.4
This release establishes the author-side mathematical core freeze of STO-OPT-MATH-001 v0.2.4, following independent mathematical and computational validation of v0.2.3 and the incorporation of the final recommended provenance and reproducibility corrections. The manuscript develops a rigorous mathematical treatment of identifiability, representation-induced masking, bounded-noise minimax recovery, covariance-weighted estimation, principal-angle sensitivity, and conditional robustness in causal phase–memory diagnostics. The principal results include: - Exact identifiability conditions and minimum linear readout dimensions. - Sharp bounded-noise minimax recovery bounds. - Sign-cone characterization of masking under weighted absolute aggregation. - Exact representation-induced collisions in the archived synthetic optical model. - Population Bayes-error identities under the verified original sampling law. - Analytical exclusion of FIR masking within the original engineered clean domain. - Explicit separation of population risk, conditional error expectations, and realized classification errors. - Quantitative documentation of noise-induced limitations and conventional estimator comparisons. The population Bayes error of approximately 7.05% is consistent with the archived clean classification error of 7.92% under the original sampling law. This agreement is conditional and does not establish causal attribution of all classification errors. The release preserves the original synthetic results and mathematical statements without retuning, classifier refitting, or new confirmatory execution. Reproducibility materials include the complete LaTeX source, immutable historical source archives, original synthetic archives, mathematical verification scripts, replay diagnostics, environment documentation, and a verified SHA-256 manifest. The original collision audit reproduces byte-identically under NumPy 2.3.5 in the documented local numerical environment. Independent mathematical and computational validation contributions by B Manvitha (India; ORCID: 0009-0006-3196-4007) are gratefully acknowledged. This release constitutes a mathematical research freeze, not experimental confirmation, formal journal peer review, or validation of a new physical law. Prospective robustness studies and experimental applications remain subject to separate qualification and execution protocols. Controlling historical synthetic results: https://doi.org/10.5281/zenodo.24308534
Authors
- Marcel Krüger (ORCID: https://orcid.org/0009-0002-5709-9729)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-19
- DOI
- https://doi.org/10.5281/zenodo.24374277
- Primary Topic
- Orbital Angular Momentum in Optics
- Type
- preprint