OctoInformBeta

OpenAI preprint, September 2026 · Section 2. Reductions and Newton potentials

Proposition 2.4: Explicit constants for ball masses and truncated potentials of positive entire solutions

Source bounds and Newton representation.

Hover over a blue or green term in the text, or a node in the graph, to find it in the other. Select it to read its declaration both in Lean and in words.

The Setting

Let n≥2n\geq2 and let p,q,A,B∈ℝp,q,A,B\in\mathbb R. Write BR={x∈ℝn:|x|<R}B_R=\{x\in\mathbb R^n:|x|<R\} and denote Lebesgue measure by dxdx. A positive solution (u,v)(u,v) is a pair of continuous, strictly positive real functions on ℝn\mathbb R^n, both of class C2C^2 off the origin, satisfying (1) at every nonzero point: −Δu(x)=|x|Av(x)p,−Δv(x)=|x|Bu(x)q(x≠0).-\Delta u(x)=|x|^A v(x)^p,\qquad -\Delta v(x)=|x|^B u(x)^q\qquad(x\ne0).(1)✦

The weighted sources f,g:ℝn→ℝf,g:\mathbb R^n\to\mathbb R of a solution are defined by (2), with all powers interpreted as real powers: f(x)={|x|Bu(x)q,x≠0,0,x=0,g(x)={|x|Av(x)p,x≠0,0,x=0.f(x)=\begin{cases}|x|^B u(x)^q,&x\ne0,\\0,&x=0,\end{cases} \qquad g(x)=\begin{cases}|x|^A v(x)^p,&x\ne0,\\0,&x=0.\end{cases}(2)✦

The scaling exponents α\alpha and β\beta of the system (1) are the real numbers in (3). Reciprocals at zero in these definitions have value zero. α=2+A+p(2+B)pq−1,β=2+B+q(2+A)pq−1.\alpha=\frac{2+A+p(2+B)}{pq-1},\qquad \beta=\frac{2+B+q(2+A)}{pq-1}.(3)

For a real function hh, write h+=max⁡(h,0)h_+=\max(h,0) pointwise. The source mass Mh(R)M_h(R) is the real ball integral in (4): Mh(R)=∫BRh(x)dx(h=f,g).M_h(R)=\int_{B_R}h(x)\,dx\qquad(h=f,g).(4)✦ The source measure μh\mu_h is the measure with density h+h_+ relative to Lebesgue measure, as in (5), where ∫+\int^+ denotes a nonnegative extended integral. Here EE is a measurable subset of ℝn\mathbb R^n. μh(E)=∫E+h+(x)dx(h=f,g).\mu_h(E)=\int_E^+h_+(x)\,dx\qquad(h=f,g).(5) Undecorated integrals have real values.

For a measure μ\mu, a real function hh, and a nonnegative extended kernel k(x,y)k(x,y), the kernel potential 𝒫μ,kh\mathcal P_{\mu,k}h is the [0,∞][0,\infty]-valued function defined by (6): 𝒫μ,kh(x)=∫+h+(y)k(x,y)dμ(y).\mathcal P_{\mu,k}h(x)=\int^+h_+(y)k(x,y)\,d\mu(y).(6)✦

The Constants

Let σn−1\sigma_{n-1} be the total unnormalized surface measure of the unit sphere induced by Lebesgue measure on ℝn\mathbb R^n.✦ The Newton normalization DnD_n is the real constant and the Newton kernel KnK_n on ℝn×ℝn\mathbb R^n\times\mathbb R^n is the real function given in (7), including the real-power convention when x=yx=y: Dn=(n−2)σn−1,Kn(x,y)=Dn−1|x−y|2−n.D_n=(n-2)\sigma_{n-1},\qquad K_n(x,y)=D_n^{-1}|x-y|^{2-n}.(7) The ball kernel constant dnd_n, the unit-ball weight mass Wn(t)W_n(t), and the local kernel mass JnJ_n are the real quantities in (8): dn=22−nDn,Wn(t)=∫B1|x|tdx(t∈ℝ),Jn=∫B12|x|2−ndx.d_n=\frac{2^{2-n}}{D_n},\qquad W_n(t)=\int_{B_1}|x|^t\,dx\quad(t\in\mathbb R),\qquad J_n=\int_{B_{12}}|x|^{2-n}\,dx.(8)

The source mass constant CsC_s of the system (1) is the real number in (9); put s=n−2−αs=n-2-\alpha for its associated growth exponent. Cs=((dnWn(A))(dnWn(B))p)1/(1−pq)dn.C_s=\frac{\bigl((d_nW_n(A))(d_nW_n(B))^p\bigr)^{1/(1-pq)}}{d_n}.(9)✦ The dyadic tail number T(C,η,s)T(C,\eta,s) is the real expression in (10), for real arguments C,η,sC,\eta,s: T(C,η,s)=2ηC2s8s−η1−2s−η.T(C,\eta,s)=2^\eta\frac{C\,2^s8^{s-\eta}}{1-2^{s-\eta}}.(10)✦ The uniform tail constant CtC_t and the local mass constant CmC_m of the system (1) are the real numbers in (11), where |B4||B_4| denotes the Lebesgue volume of B4B_4. These constants depend only on n,p,q,A,Bn,p,q,A,B. Ct=T(Cs,n−2,s),Cm=JnCs8s+Ct|B4|Dn.C_t=T(C_s,n-2,s),\qquad C_m=\frac{J_nC_s8^s+C_t|B_4|}{D_n}.(11)

The Claim

Theorem 1. Let n≥2n\geq2, let p,q,A,B∈ℝp,q,A,B\in\mathbb R, and let (u,v)(u,v) be a positive solution of (1). Assume q>0q>0. At every x≠0x\ne0, its Newton potential satisfies (12), with the right side viewed in [0,∞][0,\infty]: 𝒫dx,(Kn)+g(x)=u(x)(x≠0).\mathcal P_{dx,(K_n)_+}g(x)=u(x)\qquad(x\ne0).(12)✦ If also p>0p>0, the source mass satisfies (13) for every R>0R>0: Mg(R)≤CsRn−2−α(R>0).M_g(R)\leq C_s R^{n-2-\alpha}\qquad(R>0).(13)✦ If also p>0p>0, the local mass of uu satisfies (14): ∫B4u(x)dx≤Cm.\int_{B_4}u(x)\,dx\leq C_m.(14)✦ If also p>0p>0, then for |x|≤4|x|\leq4 and 0<ℓ≤10<\ell\leq1, the truncated potential satisfies (15): ∫{y:|x−y|≥ℓ}+|x−y|2−ndμg(y)≤[(Cs8n−2−α+Ct)ℓ2−n]+.\int_{\{y:\,|x-y|\geq\ell\}}^+|x-y|^{2-n}\,d\mu_g(y) \leq\bigl[(C_s8^{n-2-\alpha}+C_t)\ell^{2-n}\bigr]_+.(15)✦

In the paper, the result reads: for n≥3n\geq3, p,q>0p,q>0, A,B>−2A,B>-2 and pq>1pq>1, there is a universal CC such that every solution satisfies Mf(R)≤CRn−2−βM_f(R)\leq CR^{n-2-\beta} and Mg(R)≤CRn−2−αM_g(R)\leq CR^{n-2-\alpha} for every R>0R>0, the additive potential constants vanish with u(x)=∫Kn(x,y)g(y)dyu(x)=\int K_n(x,y)g(y)\,dy and v(x)=∫Kn(x,y)f(y)dyv(x)=\int K_n(x,y)f(y)\,dy off zero, and moreover ∫B4(f+g+u+v)dx≤C\int_{B_4}(f+g+u+v)\,dx\leq C and ∫{|x−y|≥ℓ}Kn(x,y)(f(y)+g(y))dy≤Cℓ2−n\int_{\{|x-y|\geq\ell\}}K_n(x,y)(f(y)+g(y))\,dy\leq C\ell^{2-n} for x∈B2x\in B_2 and 0<ℓ≤10<\ell\leq1.

What Is Checked

The claims are stated over positive solutions on Euclidean space, the real source masses in (4), and the nonnegative extended potentials in (6) with source measures from (5).