claimroutineinferred_dep
Per-branch cost and total exponent
Claim·§5.5 Constructing the forms and completing the proof·inside thmCounting zeros of sparse polynomials faster
Statement (paper)
For all sufficiently large , on every branch one may use the original precision , and both and are . Then , , and , in dimension , has exponent . All substitution data, coefficient arrays, and interpolation systems have total representation length bits, and multiplying the per-branch running time by at most leaves this exponential bound unchanged, so summing the branch answers gives the required running time.
In plain English
For large , on every branch the precision can stay at and the parameters , , , are polynomial in and , so the counting cost has exponent . All data have size bits, and summing over at most branches keeps this exponent, giving the theorem's running time.
Why it's here
This claim finishes the proof of the theorem. It inserts the concrete sizes of , , , from the parameter-verification claim into the counting cost from the counting proposition and the general routine. The branching overhead is then absorbed into the same exponent, and the polynomial space and representation bounds follow.
Classified routine:
Parameter bookkeeping: it substitutes the per-branch choices of r, Q, R, D from the preceding claim into the cost bounds of prop-count and prop-general-count. The 2^{n/sqrt(lambda)} branching factor is then absorbed into the same exponent.
Uses
Used by
Nothing in the paper
Sources
PaperOpen
- Section
- §5.5 Constructing the forms and completing the proof
- Page
- —
- Label
—- Line
main_full_version.tex:3170
Leannot aligned
Phase 2 aligns this node to a Lean declaration.
Friction
- infoinferred_depoutline: "Per-branch running time and total exponent" rests on claim-thm-overview-sparse-polynomials-2
- infoinferred_depoutline: "Per-branch running time and total exponent" rests on prop:count
- infoinferred_depoutline: "Per-branch running time and total exponent" rests on prop:general-count