By OpenAISourceGitHub

Manuscripts from openai/math

Fork
Clone
git clone https://openversions.com/git/index/openai-math.git
Cite
Version link
https://openversions.com/index/openai-math/commit/dfd8c46870b48511e697cbb43a3c8c91d375fa2f
Version identifier (commit hash)
dfd8c46870b48511e697cbb43a3c8c91d375fa2f

Fork to create your own copy with credit to the source. Clone to get a Git copy on your computer.

All papers

Claims

Findings

Slope-field perturbations of the two-cylinder covering

Every regular tetrahedron K⊂R3K \subset \mathbb{R}^3 of positive edge length admits a finite cylinder cover with compact parallelogram perpendicular bases (and also finite covers with bounded open measurable bases) such that ∑i∣Bi∣<12Amin⁡(K)\sum_i |B_i| < \frac{1}{2} A_{\min}(K) and ∑i∣Bi∣∣πui⊥K∣<12\sum_i \frac{|B_i|}{|\pi_{u_i^\perp} K|} < \frac{1}{2}.

Every nondegenerate tetrahedron T⊂R3T \subset \mathbb{R}^3 admits a finite cylinder cover C\mathcal{C} with compact parallelogram perpendicular bases (and also with bounded open measurable bases) and R(C;T)<1/2\mathcal{R}(\mathcal{C}; T) < 1/2, placing the infimum relative cost in [1/3,1/2)[1/3, 1/2).

For fixed d>0d > 0 and a positive function q(ϵ)q(\epsilon) with q(ϵ)=O(ϵ2)q(\epsilon) = O(\epsilon^2) and ϵ3=o(q(ϵ))\epsilon^3 = o(q(\epsilon)) as ϵ↓0\epsilon \downarrow 0, for all sufficiently small ϵ>0\epsilon > 0, K1⊂S+∪S−K_1 \subset S_+ \cup S_- and Cd(ϵ,q)h=18+ϵ2(Jd−Id)+q2+O(ϵ4)\frac{C_d(\epsilon, q)}{h} = \frac{1}{8} + \epsilon^2(J_d - I_d) + \frac{q}{2} + O(\epsilon^4), with lim⁡d↓0Jd=1615360\lim_{d \downarrow 0} J_d = \frac{16}{15360} and lim⁡d↓0Id=1715360\lim_{d \downarrow 0} I_d = \frac{17}{15360}.

For every 1<p<(51−5)/21 < p < (\sqrt{51} - 5)/2 and 1<γ<1/(2−p)1 < \gamma < 1/(2 - p), all sufficiently small positive cc determine a finite cylinder cover of the regular edge-2 tetrahedron K2K_2 with total perpendicular base area less than 2/2\sqrt{2}/2, consisting of two finite families with compact parallelogram bases and two cylinders with compact triangular bases; before the finite approximation errors, the area bound is 22+22κpc2+o(c2)\frac{\sqrt{2}}{2} + 2\sqrt{2}\kappa_p c^2 + o(c^2) with κp=(1/2)4(2p2+10p−13)30p<0\kappa_p = \frac{(1/2)^4(2p^2 + 10p - 13)}{30p} < 0.

The regular edge-2 tetrahedron K2K_2 admits a finite cylinder cover with compact convex parallelogram perpendicular bases whose total area is less than 12min⁡∣u∣=1∣πu⊥K2∣\frac{1}{2} \min_{|u|=1} |\pi_{u^\perp} K_2|.

Methods

Rational stream-function slope-field perturbation with quartic padding and finite square-tile realization via Theorem 1.2 of [6].

Source excerpts

“The standard stream-function form makes the divergence zero and the constant value of ψ on either wedge wall makes the field tangent to that wall. We use the straight-line maps p 7→ p + ϵxWd(p), not a flow of Wd.” · Show in paper

“Apply Theorem 1.2 of [6] with V = +ϵWd for S+, and with V = −ϵWd followed by the isometry (x, y, z) 7→ (y, x, −z) for S−.” · Show in paper

Monotone height-threshold inversion proof for fixed quadratic padding.

Source excerpts

“For fixed quadratic padding there is a second way to locate the overlap: give an explicit sufficient height for each family.” · Show in paper

Logarithmic-cell and product-cell finite realizations of the rational-field cover in orthonormal and centered coordinates.

Source excerpts

“In these coordinates the logarithmic-cell theorem gives bounded open bases; in the centered coordinates the product-cell theorem does so under a weaker smoothness assumption.” · Show in paper

Singular power-law slope-field perturbation with tip cutoff and two triangular-base tip cylinders.

Source excerpts

“We now keep a power-law singularity in the derivative and remove a small tip region instead. Two triangular-base cylinders cover the removed regions; their cost is smaller than the quadratic saving from the tilted families.” · Show in paper

Angular field with square-zero differential away from the apex combined with a finite apex velocity net.

Source excerpts

“The next section fixes a truncated label set, applies the square-zero case of the same finite-approximation theorem, and covers the apex segments by a finite velocity net.” · Show in paper