Logarithmic analytic rank with an explicit infinite boundary
DefinitionDetSupportRank_DeterminantSupport_Supplement_AnalyticRankalgebracombinatoricsdeterminantsupportp2o-batch-stoc27sp2o-gran-per-chapterp2o-plan-paperp2o-v1
For a function on a finite set and a finite field of cardinality , define
when the bias is positive, and give nonpositive bias the value . This extended-real convention prevents a finite rank bound from arising through a total logarithm at zero. On the multilinear source domain of positive degree the bias is positive; that fact is not assumed as an extra hypothesis. Field nontriviality ensures .
Definition code
import Definitions.Def_DetSupportRank_DeterminantSupport_Supplement_UniformBias
noncomputable section
open scoped BigOperators
namespace DetSupportRank.DeterminantSupport.Supplement
/-- The actual -log_q(bias), §1, p.2. Nonpositive bias is assigned +∞,
so a finite upper bound cannot come from Lean's total log at zero. -/
def analyticRank {A F : Type*} [Fintype A] [Field F] [Fintype F] (T : A → F) : EReal :=
if 0 < uniformBias T then
((-Real.log (uniformBias T) / Real.log (Fintype.card F : ℝ)) : ℝ)
else ⊤
end DetSupportRank.DeterminantSupport.Supplement
Source
Neil Krishnan, Guy Moshkovitz, Optimal partition rank bounds for the determinant and other multilinear polynomials, STOC 2027 submission #631 (October 2026), §1, definition of analytic rank, p. 2 (PDF p. 2); https://github.com/sigact-stoc/stoc2027/releases/download/papers-1/631.pdf