Uniqueness conjecture for curve shortening flow and blooming at infinity

Let (R2,g)(\mathbb{R}^2,g) be the plane equipped with a complete smooth O(2)O(2)-invariant metric with non-positive curvature. Curve shortening flow is unique on (R2,g)(\mathbb{R}^2,g) when any two uniformly proper solutions with the same initial data have the same image wherever both solutions are defined. The metric gg allows blooming at infinity if there exist T(0,)T\in(0,\infty) and a solution R:(0,T)(0,)R:(0,T)\to(0,\infty) of

Rt(t)=ϕr(R(t))\frac{\partial R}{\partial t}(t)=-\frac{\partial\phi}{\partial r}(R(t))

such that R(t)R(t)\to\infty as t0t\searrow0, where g=dr2+e2ϕ(r)dθ2g=dr^2+e^{2\phi(r)}d\theta^2 in polar coordinates; otherwise, gg does not allow blooming at infinity.

Uniqueness conjecture. CSF is unique on (R2,g)(\mathbb{R}^2,g) if and only if gg does not allow blooming at infinity.

The conjecture asserts that, within this class of complete rotationally symmetric non-positively curved metrics, blooming at infinity is precisely the obstruction to uniqueness for CSF with arbitrary initial data. The source states this tentatively and does not give a resolution; uniqueness is proved in the special case of solutions starting from a radial geodesic when blooming at infinity is absent.

Sources & referencesView supporting material

Primary source

Luke Thomas Peachey, “Non-uniqueness of curve shortening flow”, arXiv:2205.03442 (2022).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.