OctoInformBeta

OpenAI preprint, September 2026 · §2. The equation and its polynomial cylinder

Proof of Proposition 2.2: shift and polynomial inverse: For a coordinate pp, frame polynomials give inverse identities, with cancellation of p3p^3 required in one direction

Polynomial shifts and cylinder inverse identities.

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

For the coordinate ring AA and defining polynomial HH in (1), the cylinder of the complex affine hypersurface X=Spec⁡AX=\operatorname{Spec}A has coordinates (p,s,u,F,J,w)(p,s,u,F,J,w), with H=0H=0 and ww free. x=s2+u3+p2F,H=x2F−(1+2sx)J−p2J2−pu,A=ℂ[p,s,u,F,J]/(H).\begin{aligned} x&=s^2+u^3+p^2F,\\ H&=x^2F-(1+2sx)J-p^2J^2-pu,\\ A&=\mathbb C[p,s,u,F,J]/(H). \end{aligned}(1)

For a commutative ring RR, a coordinate tuple q=(p,s,u,F,J)q=(p,s,u,F,J) is an element of R5R^5.✦ The symbols x(q)x(q) and H(q)H(q) denote evaluation of (1) at qq. A parameter frame t=(p,s,u,M)t=(p,s,u,M) over a commutative ring RR is an element of R4R^4.✦ The base coordinate x0x_0 of either tuple is x0=s2+u3x_0=s^2+u^3.✦ The derived coordinate x(q)∈Rx(q)\in R is the first polynomial expression in (1), evaluated at qq. The hypersurface equation H(q)∈RH(q)\in R is the second polynomial expression in (1), evaluated at qq.

The scaled cubic difference k(p,u,w)k(p,u,w) is the polynomial in (2). For w∈Rw\in R, the polynomial shift Sw:R5→R5S_w:R^5\to R^5 is defined in (3), where x=x(q)x=x(q) and k=k(p,u,w)k=k(p,u,w). k(p,u,w)=−3u2w+3up2w2−p4w3.k(p,u,w)=-3u^2w+3up^2w^2-p^4w^3.(2)✦ Sw(q)=(p,s+p2xk,u−p2w,F−(2sx+1)k−p2x2k2,J−x2k).S_w(q)=\bigl(p,\ s+p^2xk,\ u-p^2w,\ F-(2sx+1)k-p^2x^2k^2,\ J-x^2k\bigr).(3)✦

The Frame Polynomials

For a parameter frame t=(p,s,u,M)t=(p,s,u,M), the linear polynomials ft,jt∈R[l]f_t,j_t\in R[l] are defined by (4); the coefficients are f0=(1+2sx0)Mf_0=(1+2sx_0)M✦, f1=4s2f_1=4s^2✦, j0=x02Mj_0=x_0^2M✦ and j1=2sx0−1j_1=2sx_0-1✦. The tuple qt(l)q_t(l) in that display is the associated linear family of points. ft(l)=f1l+f0,jt(l)=j1l+j0,qt(l)=(p,s,u,ft(l),jt(l)).\begin{aligned} f_t(l)&=f_1l+f_0, & j_t(l)&=j_1l+j_0,\\ q_t(l)&=(p,s,u,f_t(l),j_t(l)). \end{aligned}(4)

The quadratic correction Qt∈R[l]Q_t\in R[l] is defined by (5). For a parameter frame tt, the constant correction ct∈Rc_t\in R is defined by ct=Qt(0)c_t=Q_t(0).✦ Qt(l)=2x0ft(l)2−2sft(l)jt(l)−jt(l)2.Q_t(l)=2x_0f_t(l)^2-2sf_t(l)j_t(l)-j_t(l)^2.(5)✦

Write q1,t,q2,t∈Rq_{1,t},q_{2,t}\in R for the coefficients of the correction polynomial Q1,t∈R[l]Q_{1,t}\in R[l] in (6). q1,t=4x0f1f0−2s(f1j0+f0j1)−2j1j0,q2,t=2x0f12−2sf1j1−j12,Q1,t(l)=q1,t+q2,tl.\begin{aligned} q_{1,t}&=4x_0f_1f_0-2s(f_1j_0+f_0j_1)-2j_1j_0,\\ q_{2,t}&=2x_0f_1^2-2sf_1j_1-j_1^2,\\ Q_{1,t}(l)&=q_{1,t}+q_{2,t}l. \end{aligned}(6)

Equation (7) gives the lifting base point L*,t∈RL_{*,t}\in R and lifted parameter Lt∈R[z]L_t\in R[z]. The shift parameter Wt∈R[z]W_t\in R[z] is defined in (8). L*,t=pu−p2ct,Lt(z)=L*,t+p3z.L_{*,t}=pu-p^2c_t,\qquad L_t(z)=L_{*,t}+p^3z.(7) Wt(z)=−z−(u−pct+p2z)Q1,t(Lt(z))−pft(Lt(z))3.W_t(z)=-z-(u-pc_t+p^2z)Q_{1,t}(L_t(z))-pf_t(L_t(z))^3.(8)✦

Write ht∈R[l,w]h_t\in R[l,w] and et∈R[l,w]e_t\in R[l,w] for the residual and encoding coordinates in (9). ht(l,w)=u−pQt(l)−p3ft(l)3−p2w,et(l,w)=−w−pft(l)3−Q1,t(l)ht(l,w).\begin{aligned} h_t(l,w)&=u-pQ_t(l)-p^3f_t(l)^3-p^2w,\\ e_t(l,w)&=-w-pf_t(l)^3-Q_{1,t}(l)h_t(l,w). \end{aligned}(9)

The Two Parametrizations

A coordinate tuple q=(p,s,u,F,J)q=(p,s,u,F,J) determines a linear parameter l(q)∈Rl(q)\in R and an associated frame t(q)∈R4t(q)\in R^4 by (10). Here M(q)M(q) is the fourth coordinate of the frame. l(q)=x02F−(1+2sx0)J,M(q)=(1−2sx0)F+4s2J,t(q)=(p,s,u,M(q)).\begin{aligned} l(q)&=x_0^2F-(1+2sx_0)J,\\ M(q)&=(1-2sx_0)F+4s^2J,\\ t(q)&=(p,s,u,M(q)). \end{aligned}(10)

Equation (11) defines the polynomial decoding map 𝒟:R4×R→R5×R\mathcal D:R^4\times R\to R^5\times R and encoding map ℰ:R5×R→R4×R\mathcal E:R^5\times R\to R^4\times R. For the encoding formula, put r=S−w(q)r=S_{-w}(q). 𝒟(t,z)=(SWt(z)(qt(Lt(z))),Wt(z)),ℰ(q,w)=(t(r),et(r)(l(r),w)).\begin{aligned} \mathcal D(t,z)&=\bigl(S_{W_t(z)}(q_t(L_t(z))),\ W_t(z)\bigr),\\ \mathcal E(q,w)&=\bigl(t(r),\ e_{t(r)}(l(r),w)\bigr). \end{aligned}(11)

The Claim

Theorem 1. Let RR be a commutative ring. For a coordinate tuple qq and w∈Rw\in R, the polynomial shift satisfies (12). H(Sw(q))=H(q)+p3w.H(S_w(q))=H(q)+p^3w.(12)✦ For a parameter frame tt and l,w∈Rl,w\in R, the encoding coordinate has the certificate (13). p3et(l,w)−(l−L*,t)=(p2Q1,t(l)−1)(H(qt(l))+p3w).p^3e_t(l,w)-(l-L_{*,t}) =(p^2Q_{1,t}(l)-1)\bigl(H(q_t(l))+p^3w\bigr).(13)✦ For a parameter frame tt and z∈Rz\in R, the coordinate equation is (14), and encoding after decoding satisfies (15). H(qt(Lt(z)))+p3Wt(z)=0.H(q_t(L_t(z)))+p^3W_t(z)=0.(14)✦ ℰ(𝒟(t,z))=(t,z).\mathcal E(\mathcal D(t,z))=(t,z).(15)✦ If H(q)=0H(q)=0 and multiplication by p3p^3 is injective on RR, decoding after encoding satisfies (16) for every w∈Rw\in R. 𝒟(ℰ(q,w))=(q,w).\mathcal D(\mathcal E(q,w))=(q,w).(16)✦

In the paper, the result reads: the exponential shift changes HH to H+p3wH+p^3w, and, with linear coordinates l,Ml,M, root L*,tL_{*,t} and polynomial ete_t, the certificate p3et(l,w)−(l−L*,t)=(p2Q1,t(l)−1)(H(qt(l))+p3w)p^3e_t(l,w)-(l-L_{*,t})=(p^2Q_{1,t}(l)-1)(H(q_t(l))+p^3w) gives polynomial coordinates and an inverse without inverting pp.

What Is Checked

The statements concern coordinate tuples and parameter frames over an arbitrary commutative ring RR, with injectivity of multiplication by p3p^3 required only in (16).