OctoInformBeta

OpenAI preprint, September 2026 · §5. Uniform control of returning geodesics

Lemma 5.4 [Coefficient estimates replacing the S bounds]: Explicit scalar estimates for either boundary end under smallness and cometric hypotheses

Coefficient bounds control returning geodesic energy.

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

A returning neck geodesic has inward height τ(s)\tau(s) equal to zero at both ends of its time interval [0,L][0,L]. Its conserved transverse momentum is p∈ℝ2p\in\mathbb R^2, and its height velocity is v(s)v(s). The library states scalar coefficient estimates and an energy inequality involving the transverse endpoint displacement.

The transverse coordinate plane is ℝ2\mathbb R^2, and transverse matrices are real 2×22\times2 matrices. For a real 2×22\times2 matrix BB and vectors u,v,p∈ℝ2u,v,p\in\mathbb R^2, the matrix bilinear pairing ℬ(B;u,v)\mathcal B(B;u,v), the quadratic form qB(p)q_B(p), and the squared Euclidean norm |p|2|p|^2 are real quantities defined by ℬ(B;u,v)=u𝖳Bv,qB(p)=ℬ(B;p,p),|p|2=p12+p22.\mathcal B(B;u,v)=u^{\mathsf T}Bv,\qquad q_B(p)=\mathcal B(B;p,p),\qquad |p|^2=p_1^2+p_2^2.(1) The tangential basis vector is e=(1,0)e=(1,0), with coordinates numbered as in (1).✦

The one-sided cutoff aa and its integrated cutoff AA are real functions defined by a(u)={exp⁡(−1/(2/3−u)),u<2/3,0,u≥2/3,A(u)=∫0ua(y)dy.a(u)= \begin{cases} \exp\!\bigl(-1/(2/3-u)\bigr),&u<2/3,\\ 0,&u\geq2/3, \end{cases} \qquad A(u)=\int_0^u a(y)\,dy.(2) Let θ\theta denote the standard smooth transition from zero to one. The shear transition χ\chi, cometric shear σ\sigma, and diagonal coefficients d1,d2d_1,d_2 at neck height tt and parameter δ\delta are the real functions χ(t)=θ(6(t−5/12)),σ(δ,t)=δ2χ(t),d1(δ,t)=1+δA(t),d2(δ,t)=1+δA(1−t).\begin{aligned} \chi(t)&=\theta\bigl(6(t-5/12)\bigr),& \sigma(\delta,t)&=\delta^2\chi(t),\\ d_1(\delta,t)&=1+\delta A(t),& d_2(\delta,t)&=1+\delta A(1-t). \end{aligned}(3) The cometric matrix P(δ,t)P(\delta,t) is the real matrix formed from (3) by P(δ,t)=(d1(δ,t)+d2(δ,t)σ(δ,t)2d2(δ,t)σ(δ,t)d2(δ,t)σ(δ,t)d2(δ,t)).P(\delta,t)= \begin{pmatrix} d_1(\delta,t)+d_2(\delta,t)\sigma(\delta,t)^2& d_2(\delta,t)\sigma(\delta,t)\\ d_2(\delta,t)\sigma(\delta,t)&d_2(\delta,t) \end{pmatrix}.(4)✦ The endpoint frame matrix C2(δ)C_2(\delta) and terminal cometric Pend(δ,τ)P_{\mathrm{end}}(\delta,\tau) are C2(δ)=(011−δ2),Pend(δ,τ)=C2(δ)P(δ,1−τ)C2(δ)𝖳.C_2(\delta)=\begin{pmatrix}0&1\\1&-\delta^2\end{pmatrix},\qquad P_{\mathrm{end}}(\delta,\tau) =C_2(\delta)P(\delta,1-\tau)C_2(\delta)^{\mathsf T}.(5) For a choice rr of boundary end, the coordinate cometric Pr,δ(τ)P_{r,\delta}(\tau) is P(δ,τ)P(\delta,\tau) at the left end and Pend(δ,τ)P_{\mathrm{end}}(\delta,\tau) from (5) at the right end.✦

The Coordinate Energy

For conserved momentum pp, the momentum potential VpV_p, reduced force FpF_p, and force derivative Fp′F'_p are the real functions of inward height Vp(τ)=ℬ(Pr,δ(τ);p,p),Fp(τ)=ℬ(Pr,δ′(τ);p,p),Fp′(τ)=ℬ(Pr,δ″(τ);p,p).V_p(\tau)=\mathcal B(P_{r,\delta}(\tau);p,p),\qquad F_p(\tau)=\mathcal B(P'_{r,\delta}(\tau);p,p),\qquad F'_p(\tau)=\mathcal B(P''_{r,\delta}(\tau);p,p).(6) Primes on the cometric in (6) denote height derivatives; primes on paths below denote geodesic time derivatives.

For a transverse matrix path QQ, vector paths kk and zz, and scalar functions hh and ww on [0,L][0,L], the completed-square energy density ℰ\mathscr E is a real function on [0,L][0,L], and the cometric coordinate energy E[Q,k,h;z,w]E[Q,k,h;z,w] is its integral, defined by ℰ(s)=ℬ(Q(s);z′(s)−w(s)k(s),z′(s)−w(s)k(s))+w′(s)2+h(s)w(s)2,E[Q,k,h;z,w]=∫0Lℰ(s)ds.\begin{aligned} \mathscr E(s)&=\mathcal B(Q(s);z'(s)-w(s)k(s),z'(s)-w(s)k(s)) +w'(s)^2+h(s)w(s)^2,\\ E[Q,k,h;z,w]&=\int_0^L\mathscr E(s)\,ds. \end{aligned}(7)

The turning constants a0a_0 and β\beta are real numbers, and the mixed coefficient bound mm is a real function of a real parameter CC, defined by a0=e−3/2,β=1+e−9/262,m(C)=a(0)+2Ca(0)+22Ca(0)/3.a_0=e^{-3/2},\qquad \beta=\frac{1+e^{-9/26}}2,\qquad m(C)=a(0)+\sqrt{2Ca(0)}+2\sqrt{2Ca(0)/3}.(8) For real coefficients Ce,CmC_e,C_m, the set 𝒟(Ce,Cm)\mathcal D(C_e,C_m) of neck parameters is the subset of ℝ\mathbb R defined by 𝒟(Ce,Cm)={δ∈ℝ:0<δ≤1,δ≤12Cm,δ≤1−β2Cm+Cea0}.\mathcal D(C_e,C_m)= \left\{\delta\in\mathbb R: 0<\delta\leq1,\quad \delta\leq\frac1{2C_m},\quad \delta\leq\frac{1-\beta}{2C_m+C_ea_0}\right\}.(9)✦

The Claim

Theorem 1. Let L≥0L\geq0, let v0v_0 be real, and let F,vF,v be continuous real functions on [0,L][0,L] satisfying F(s)+v(s)2=1F(s)+v(s)^2=1 and ∫0Lv(s)2ds≤Lv02/3\int_0^L v(s)^2\,ds\leq Lv_0^2/3. The weighted integral bound is (10): ∫0LF(s)ds+4∫0Lv(s)2ds≤L(1+v02).\int_0^L F(s)\,ds+4\int_0^L v(s)^2\,ds\leq L(1+v_0^2).(10)✦

For L≥0L\geq0, C≥0C\geq0, and δ∈[0,1]\delta\in[0,1], let ff be continuous on [0,L][0,L] with f(s)≤Cf(s)\leq C throughout that interval. The second coefficient bound is (11): ∫0Lf(s)ds+2CδL≤3CL.\int_0^L f(s)\,ds+2C\delta L\leq3CL.(11)✦

Fix either boundary end. Let δ≥0\delta\geq0, C>0C>0, and T∈[0,1]T\in[0,1]. Suppose −Pr,δ″(τ)-P''_{r,\delta}(\tau) is positive semidefinite and |Pr,δ′(τ)e|2≤C2δ2|P'_{r,\delta}(\tau)e|^2\leq C^2\delta^2 for every τ∈[0,1]\tau\in[0,1]. For a momentum pp with Fp(τ)≥0F_p(\tau)\geq0 on [0,T][0,T], the mixed cometric bound (12) holds: |ℬ(Pr,δ(τ);e,p)−p1|≤(a(0)+2Ca(0))δ|p1|,0≤τ≤T.\left|\mathcal B(P_{r,\delta}(\tau);e,p)-p_1\right| \leq\left(a(0)+\sqrt{2Ca(0)}\right)\delta|p_1|, \qquad 0\leq\tau\leq T.(12)✦

Let L≥0L\geq0 and let C,δ,p1C,\delta,p_1 be real. For a continuous real function ff on [0,L][0,L] satisfying |f(s)−p1|≤Cδ|p1||f(s)-p_1|\leq C\delta|p_1|, the mixed integral bound is (13): |∫0Lf(s)ds−Lp1|≤CδL|p1|.\left|\int_0^L f(s)\,ds-Lp_1\right|\leq C\delta L|p_1|.(13)✦

For the returning energy bound, fix either boundary end, C≥18C\geq18, a momentum pp, and δ∈𝒟(3C,m(C))\delta\in\mathcal D(3C,m(C)). Assume the cometric estimates (14) for every τ∈[0,1]\tau\in[0,1] and q∈ℝ2q\in\mathbb R^2: |q|29≤qPr,δ(τ)(q)≤18|q|2,|Pr,δ′(τ)e|2≤C2δ2,q−Pr,δ″(τ)(q)≥0,|Pr,δ′(τ)p|2≤Cδq−Pr,δ″(τ)(p).\begin{aligned} \frac{|q|^2}{9}&\leq q_{P_{r,\delta}(\tau)}(q)\leq18|q|^2,\\ |P'_{r,\delta}(\tau)e|^2&\leq C^2\delta^2,\qquad q_{-P''_{r,\delta}(\tau)}(q)\geq0,\\ |P'_{r,\delta}(\tau)p|^2&\leq C\delta\,q_{-P''_{r,\delta}(\tau)}(p). \end{aligned}(14) Let L>0L>0 and let τ,v\tau,v be differentiable on [0,L][0,L], satisfying (15) there, with 0<τ(s)<10<\tau(s)<1 for 0<s<L0<s<L: τ′(s)=v(s),v′(s)=−12Fp(τ(s)),Vp(τ(s))+v(s)2=1,τ(0)=τ(L)=0,τ(L/2)>14.\begin{gathered} \tau'(s)=v(s),\qquad v'(s)=-\tfrac12F_p(\tau(s)),\qquad V_p(\tau(s))+v(s)^2=1,\\ \tau(0)=\tau(L)=0,\qquad \tau(L/2)>\tfrac14. \end{gathered}(15) Let z:[0,L]→ℝ2z:[0,L]\to\mathbb R^2 and w:[0,L]→ℝw:[0,L]\to\mathbb R be differentiable with continuous derivatives and w(0)=w(L)=0w(0)=w(L)=0. In (7), take Q(s)=Pr,δ(τ(s))−1Q(s)=P_{r,\delta}(\tau(s))^{-1}, k(s)=Pr,δ′(τ(s))pk(s)=P'_{r,\delta}(\tau(s))p, and h(s)=−Fp′(τ(s))/2h(s)=-F'_p(\tau(s))/2. Writing Δz=z(L)−z(0)\Delta z=z(L)-z(0), the returning energy bound is (16): 0≤E[Q,k,h;z,w]+δa0v(0)4(e⋅Δz)2−(p⋅Δz)2L.0\leq E[Q,k,h;z,w] +\frac{\delta a_0v(0)}4(e\cdot\Delta z)^2 -\frac{(p\cdot\Delta z)^2}{L}.(16)✦

In the paper, the result reads: for the Hilbert dual form SS there are Ce,Cm>0C_e,C_m>0, uniform for sufficiently small δ>0\delta>0, such that every returning passage satisfies S(p,p)≤L(1+v02)S(p,p)\leq L(1+v_0^2), S(e,e)≤CeLS(e,e)\leq C_eL, and |S(p,e)−Lp1|≤CmδL|p1||S(p,e)-Lp_1|\leq C_m\delta L|p_1|, with constants depending only on the cometric family and cutoff, independently of p,L,v0,t*p,L,v_0,t_* and the boundary end, where v0v_0 is the initial height velocity and t*t_* is the turning height.

What Is Checked

The statements concern continuous scalar functions on compact time intervals and the explicit endpoint cometric family, with the returning energy bound stated under (9) and (14) for the height trajectories in (15) and differentiable variations with continuous derivatives whose height component vanishes at both endpoints.