Treating exact interpolation as numerically benign

Exact statement

Treating exact interpolation as numerically benign. integer nodes $1,\ldots,m$ give exact weights $(-1)^{j-1}\binom mj$ whose absolute sum is $2^m-1$ . They are convenient for rational height but can amplify floating-point error. Roots of unity improve the interpolation weights but change the field and do not control within-decomposition cancellation.

Statusrejected.

External reviewNo documented external or specialist review of the FPRD synthesis is recorded.

Context

The research record preserves failed approaches when they expose a model boundary, an invalid transfer, or an unsupported strengthening.

Proof or evidence

integer nodes $1,\ldots,m$ give exact weights $(-1)^{j-1}\binom mj$ whose absolute sum is $2^m-1$ . They are convenient for rational height but can amplify floating-point error. Roots of unity improve the interpolation weights but change the field and do not control within-decomposition cancellation.

Rejected transfers and surviving boundaries →

Verification notes

The failure status and preserved lesson are recorded in the cumulative failure ledger.

Limitations

exact coefficient height and numerical conditioning are different receipt coordinates.

Open work

Keep this exclusion attached to any future reuse of the corresponding construction.