← 研踪 PaperTrailarXiv:2602.07834 · v3
What Does Machine Learning Learn from Numerical Calabi--Yau Geometry? A Controlled Study on the Fermat Quintic
首次提交 2026-02-08 13:51
版本更新 2026-11-16 18:59
雷达观测 2026-11-17 17:11
摘要
We study polynomial regression of numerical metric errors on the Fermat quintic. A known hypersurface volume-ratio identity gives the exact reference relative to the induced Fubini--Study metric, where on unit-normalized homogeneous coordinates. We optimize algebraic metrics at degrees while preserving finite Fermat symmetry, then fit their remaining logarithmic volume-form error. Regression splits and bootstrap draws keep the five roots of each sampled line together, with all compared teachers evaluated on common clouds. Warm-start replication correlations exceed at . Across degrees, source-derived modes fall short of target-specific fits, including in a post-hoc quartic check, but outperform random spans of matched rank in the tested libraries. Two independent random-invariant starts have residual correlations --, while their fitted cubic modes correlate at -- and retain useful calibrated cross-start prediction. Their fitted modes agree poorly with those of the principal warm-start teacher. Adding the phase-sensitive invariant improves prediction beyond the cubic modulus-only library for all seven warm-start teachers. The gain depends on the library: at , a quartic modulus-only fit explains about 97--98\% of residual variance and the conditional interval for the added gain includes zero for both replicas under both measures. The gains therefore establish neither phase independence nor irreducible phase necessity; fixed-baseline augmentation preserves this distinction. The fitted components depend on the numerical protocol; global optimality and a universal finite-degree correction remain unestablished.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v3 · 2026-11-16 18:59 · 最新观测版本
What Does Machine Learning Learn from Numerical Calabi--Yau Geometry? A Controlled Study on the Fermat Quintic
We study polynomial regression of numerical metric errors on the Fermat quintic. A known hypersurface volume-ratio identity gives the exact reference relative to the induced Fubini--Study metric, where on unit-normalized homogeneous coordinates. We optimize algebraic metrics at degrees while preserving finite Fermat symmetry, then fit their remaining logarithmic volume-form error. Regression splits and bootstrap draws keep the five roots of each sampled line together, with all compared teachers evaluated on common clouds. Warm-start replication correlations exceed at . Across degrees, source-derived modes fall short of target-specific fits, including in a post-hoc quartic check, but outperform random spans of matched rank in the tested libraries. Two independent random-invariant starts have residual correlations --, while their fitted cubic modes correlate at -- and retain useful calibrated cross-start prediction. Their fitted modes agree poorly with those of the principal warm-start teacher. Adding the phase-sensitive invariant improves prediction beyond the cubic modulus-only library for all seven warm-start teachers. The gain depends on the library: at , a quartic modulus-only fit explains about 97--98\% of residual variance and the conditional interval for the added gain includes zero for both replicas under both measures. The gains therefore establish neither phase independence nor irreducible phase necessity; fixed-baseline augmentation preserves this distinction. The fitted components depend on the numerical protocol; global optimality and a universal finite-degree correction remain unestablished.
查看此版本解读 →查看此版本原文 ↗