Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Section 2 — counting and mixed-degree comparisons

Proved
PhilipponMultiplicity.corollary_counting_estimates

by tomasz · Sep 23, 2026 · Mathlib 0df444a (Lean v4.33.1)

draft-statementphilippon-multiplicity

Compiled open theorem statement; proof not yet supplied. Checked locally with Lean 4.33.1 and the proposal’s pinned Mathlib. An independent blind readback is attached.

The numerical and geometric comparisons in the corollary proofs: binomial lower bound, real-grid coset count from the quotient rank, mixed-degree lower bound, and a uniform degree-ratio estimate for disjoint factors.

Preamble
import Definitions.Def_PhilipponMultiplicity_GeometricSupport
set_option autoImplicit false
open scoped BigOperators
Formal statement
namespace PhilipponMultiplicity

theorem corollary_counting_estimates
    (K : Type*) [NontriviallyNormedField K] (hK : IsPhilipponBaseField K)
    (G : EmbeddedGroupProduct K) :
    (∀ (A : AnalyticSubgroup G) (H : AlgebraicSubgroup G) (T : ℕ),
      (((T + 1 : ℕ) : ℝ) ^ analyticCodimension A H.carrier) /
          (Nat.factorial (analyticCodimension A H.carrier) : ℝ) ≤
        (Nat.choose (T + analyticCodimension A H.carrier)
          (analyticCodimension A H.carrier) : ℝ)) ∧
    (∀ (A : AnalyticSubgroup G) (H : AlgebraicSubgroup G),
      ¬ A.carrier ⊆ H.carrier →
      ∀ (m : ℕ) (γ : Fin m → G.Point) (S : ℝ), 0 ≤ S →
        S ^ samplingQuotientRank γ H ≤
          (((fun x => translate x H.carrier) '' samplingGrid γ S).ncard : ℝ)) ∧
    (∀ (H : AlgebraicSubgroup G) (r : SourceMixedCodimensionIndex G H)
        (D : G.FactorIndex → ℕ), (∀ i, 1 ≤ D i) →
      (mixedDegree G H.carrier r.complementIndex : ℝ) *
        (Nat.factorial (varietyDimension G H.carrier) : ℝ) /
        (∏ i, (Nat.factorial (r.complementIndex i) : ℝ)) *
        (∏ i, (D i : ℝ) ^ r.complementIndex i) ≤ hilbertDegreeForm G H.carrier D) ∧
    (HasDisjointFactors G → ∃ c : ℝ, 0 < c ∧
      ∀ (H : AlgebraicSubgroup G) (D : G.FactorIndex → ℕ), (∀ i, 1 ≤ D i) →
        0 < hilbertDegreeForm G H.carrier D ∧
        hilbertDegreeForm G Set.univ D / hilbertDegreeForm G H.carrier D ≤
          c * ∏ i, (D i : ℝ) ^ factorCodimension G H i) := by sorry

end PhilipponMultiplicity
Source
1986, pp.360–361. https://numdam.org/articles/10.24033/bsmf.2060/
Read-back

What the Lean code literally says, in plain math · gpt-6

Let KKK be any nontrivially normed field for which there is either an isometric ring isomorphism K≅CK\cong\mathbb CK≅C, or a prime natural number ppp and an isometric ring isomorphism K≅PadicComplex⁡(p)K\cong\operatorname{PadicComplex}(p)K≅PadicComplex(p). An embedded group product GGG consists of a positive finite number sss of commutative groups EiE_iEi​, each carried by a locally closed subset of PNi(K)\mathbb P^{N_i}(K)PNi​(K), with addition and negation locally represented by nonsimultaneously-zero homogeneous polynomial tuples of a common multidegree; its points are the tuples of factor points, with componentwise addition, embedded in M=∏iPNi(K)M=\prod_i\mathbb P^{N_i}(K)M=∏i​PNi​(K). Write R=K[Xij]R=K[X_{ij}]R=K[Xij​] for the block coordinate ring, evaluate polynomials at the chosen homogeneous representatives of embedded points, and use the induced Zariski topology, where the ambient topology is generated by nonvanishing sets of block-homogeneous polynomials. A polynomial is homogeneous of multidegree DDD when every supported monomial has degree DiD_iDi​ in block iii; zero is allowed at every multidegree. An algebraic subgroup means an additive subgroup closed in this topology, and g+V={g+x:x∈V}g+V=\{g+x:x\in V\}g+V={g+x:x∈V}. For V⊆GV\subseteq GV⊆G, let I(V)\mathcal I(V)I(V) be the ideal spanned by all homogeneous polynomials vanishing on its embedded points, and let HVH_VHV​ be the rational polynomial whose values, for all coordinatewise sufficiently large natural multidegrees, are the dimensions of the images of those homogeneous pieces in R/I(V)R/\mathcal I(V)R/I(V), chosen if it exists and set to zero if none exists. Write δ(V)=deg⁡totHV\delta(V)=\deg_{\mathrm{tot}}H_Vδ(V)=degtot​HV​ (zero for the zero polynomial), deg⁡V(D)=δ(V)! (HV)δ(V)(D)∈R\deg_V(D)=\delta(V)!\,(H_V)_{\delta(V)}(D)\in\mathbb RdegV​(D)=δ(V)!(HV​)δ(V)​(D)∈R, and μV(α)=[tα]HV∏iαi!∈R\mu_V(\alpha)=[t^\alpha]H_V\prod_i\alpha_i!\in\mathbb RμV​(α)=[tα]HV​∏i​αi​!∈R if ∑iαi=δ(V)\sum_i\alpha_i=\delta(V)∑i​αi​=δ(V), with μV(α)=0\mu_V(\alpha)=0μV​(α)=0 otherwise. Here D,αD,\alphaD,α are natural multidegrees and (HV)δ(V)(H_V)_{\delta(V)}(HV​)δ(V)​ is the top total-homogeneous component. For each factor, did_idi​ is the same Hilbert-polynomial total degree computed for its carrier in its single projective space, and n=∑idin=\sum_i d_in=∑i​di​ is the defined dimension of GGG. An analytic subgroup datum AAA has a positive natural parameter dimension qqq, a positive real radius ρ\rhoρ, the open ball B(0,ρ)⊂KqB(0,\rho)\subset K^qB(0,ρ)⊂Kq, and a map a:B(0,ρ)→Ga:B(0,\rho)\to Ga:B(0,ρ)→G with a(0)=0a(0)=0a(0)=0 and a(u+v)=a(u)+a(v)a(u+v)=a(u)+a(v)a(u+v)=a(u)+a(v) whenever u,v,u+vu,v,u+vu,v,u+v lie in that ball. It also supplies functions ℓg:Kq→K{(i,j)}\ell_g:K^q\to K^{\{(i,j)\}}ℓg​:Kq→K{(i,j)} for every g∈Gg\in Gg∈G: every coordinate is analytic at zero, the coordinates of ℓ0\ell_0ℓ0​ have convergent power series on the whole ball and give nonsimultaneously-zero projective representatives of a(z)a(z)a(z) there, and, as germs at zero, the nonzero block tuples of ℓg(z)\ell_g(z)ℓg​(z) represent g+a(z)g+a(z)g+a(z). No injectivity of aaa is required. The carrier A∗A^\astA∗ is the additive subgroup generated by a(B(0,ρ))a(B(0,\rho))a(B(0,ρ)), not a topological closure. For V⊆GV\subseteq GV⊆G, put TA(V)=⋂P∈I(V)ker⁡d0(z↦P(ℓ0(z)))T_A(V)=\bigcap_{P\in\mathcal I(V)}\ker d_0(z\mapsto P(\ell_0(z)))TA​(V)=⋂P∈I(V)​kerd0​(z↦P(ℓ0​(z))) and κA(V)=q−dim⁡KTA(V)\kappa_A(V)=q-\dim_K T_A(V)κA​(V)=q−dimK​TA​(V), using natural subtraction; the defined dimension of AAA is κA({0})\kappa_A(\{0\})κA​({0}). Write νA(P,g)=inf⁡{j∈N:Dj(z↦P(ℓg(z)))(0)≠0}\nu_A(P,g)=\inf\{j\in\mathbb N:D^j(z\mapsto P(\ell_g(z)))(0)\ne0\}νA​(P,g)=inf{j∈N:Dj(z↦P(ℓg​(z)))(0)=0} in N∪{∞}\mathbb N\cup\{\infty\}N∪{∞}, where derivatives are full iterated Fréchet derivatives, order zero is included, and the infimum of an empty set is ∞\infty∞. A mixed-codimension index for HHH is a vector r∈Nsr\in\mathbb N^sr∈Ns with ri≤dir_i\le d_iri​≤di​ and ∑iri+δ(H)=n\sum_i r_i+\delta(H)=n∑i​ri​+δ(H)=n; its complementary vector is βi=di−ri\beta_i=d_i-r_iβi​=di​−ri​, using natural subtraction, and its degree monomial is ∏iDiri\prod_i D_i^{r_i}∏i​Diri​​. For a tuple γ=(γ0,…,γm−1)\gamma=(\gamma_0,\ldots,\gamma_{m-1})γ=(γ0​,…,γm−1​) with m∈Nm\in\mathbb Nm∈N and a real SSS, put Λγ(S)={∑j<majγj:aj∈N, aj≤S for every j}\Lambda_\gamma(S)=\{\sum_{j<m}a_j\gamma_j:a_j\in\mathbb N,\ a_j\le S\text{ for every }j\}Λγ​(S)={∑j<m​aj​γj​:aj​∈N, aj​≤S for every j} and let rγ(H)r_\gamma(H)rγ​(H) be the natural-number Z\mathbb ZZ-module rank, defined by finrank⁡Z\operatorname{finrank}_{\mathbb Z}finrankZ​, of the submodule spanned by the classes of γj\gamma_jγj​ in the additive quotient G/HG/HG/H. For a finite subset Σ⊆G\Sigma\subseteq GΣ⊆G, let NΣ(H)N_\Sigma(H)NΣ​(H) be the number of distinct translated sets x+Hx+Hx+H as xxx ranges over Σ\SigmaΣ; repeated coincident translates are counted once. Let Σ(j)\Sigma^{(j)}Σ(j) be the finite set of sums of exactly jjj members of Σ\SigmaΣ, with repetition allowed, so Σ(0)={0}\Sigma^{(0)}=\{0\}Σ(0)={0} even if Σ\SigmaΣ is empty. For every GGG, the theorem asserts four clauses. For every analytic datum AAA, every algebraic subgroup HHH, and every T∈NT\in\mathbb NT∈N, (T+1)κA(H)κA(H)!≤(T+κA(H)κA(H))\frac{(T+1)^{\kappa_A(H)}}{\kappa_A(H)!}\le\binom{T+\kappa_A(H)}{\kappa_A(H)}κA​(H)!(T+1)κA​(H)​≤(κA​(H)T+κA​(H)​), with both sides regarded as real numbers. For every A,HA,HA,H such that A∗⊈HA^\ast\nsubseteq HA∗⊈H, every natural mmm, every tuple γ\gammaγ of length mmm, and every real S≥0S\ge0S≥0, one has Srγ(H)≤∣{x+H:x∈Λγ(S)}∣S^{r_\gamma(H)}\le|\{x+H:x\in\Lambda_\gamma(S)\}|Srγ​(H)≤∣{x+H:x∈Λγ​(S)}∣, the right side being the natural cardinality of that set regarded as a real number (the underlying total natural-cardinality function is zero on infinite sets). For every algebraic subgroup HHH, every mixed-codimension index rrr for HHH, and every natural degree vector DDD with all Di≥1D_i\ge1Di​≥1, one has μH(β)δ(H)!∏iβi!∏iDiβi≤deg⁡H(D)\mu_H(\beta)\frac{\delta(H)!}{\prod_i\beta_i!}\prod_i D_i^{\beta_i}\le\deg_H(D)μH​(β)∏i​βi​!δ(H)!​∏i​Diβi​​≤degH​(D). Finally, suppose GGG has the following disjoint-factor property: for every algebraic subgroup HHH and every tuple x∈Gx\in Gx∈G, x∈Hx\in Hx∈H if and only if xi∈πi(H)x_i\in\pi_i(H)xi​∈πi​(H) for every factor, and every projection πi(H)\pi_i(H)πi​(H) is closed in the projective Zariski topology induced on the factor group EiE_iEi​. Then there exists a real constant c>0c>0c>0 such that for every algebraic subgroup HHH and every natural degree vector DDD with all Di≥1D_i\ge1Di​≥1, both deg⁡H(D)>0\deg_H(D)>0degH​(D)>0 and deg⁡G(D)/deg⁡H(D)≤c∏iDifi(H)\deg_G(D)/\deg_H(D)\le c\prod_iD_i^{f_i(H)}degG​(D)/degH​(D)≤c∏i​Difi​(H)​, where fi(H)=di−δi(πi(H))f_i(H)=d_i-\delta_i(\pi_i(H))fi​(H)=di​−δi​(πi​(H)) uses natural subtraction and δi\delta_iδi​ is the Hilbert-polynomial total degree of the homogeneous vanishing ideal of that projected subset in its single factor's projective space. This constant is chosen before H,DH,DH,D. No connectedness or positive dimension is assumed; T=0T=0T=0, κA(H)=0\kappa_A(H)=0κA​(H)=0, an empty tuple m=0m=0m=0, and S=0S=0S=0 are allowed, with 00=10^0=100=1 and 0!=10!=10!=1. The second clause explicitly assumes A∗⊈HA^\ast\nsubseteq HA∗⊈H, the third does not assert existence of a mixed-codimension index when none has been supplied, and the fourth explicitly includes positivity of its denominator degree.

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me