← All problems

Quadratic degree growth for three-term Gale–Robinson recurrences

Let 1≤p≤q≤r1\le p\le q\le r be integers, set S=p+q+rS=p+q+r, and consider the three-term Gale–Robinson recurrence xnxn+S=xn+pxn+q+r+xn+qxn+p+r+xn+rxn+p+qx_nx_{n+S}=x_{n+p}x_{n+q+r}+x_{n+q}x_{n+p+r}+x_{n+r}x_{n+p+q}. Let Φp,q,r\Phi_{p,q,r} be the rational map induced by this recurrence on the state vector (xn,…,xn+S−1)(x_n,\ldots,x_{n+S-1}), and let dn=deg⁡(Φp,q,rn)d_n=\deg(\Phi_{p,q,r}^n). The conjecture is that, for every such triple (p,q,r)(p,q,r), the degree sequence has quadratic growth with a periodic correction: there exist an integer T≥1T\ge1 and quadratic polynomials Q0,…,QT−1Q_0,\ldots,Q_{T-1} such that dn=Qn mod T(n)d_n=Q_{n\bmod T}(n) for all sufficiently large nn.

References

Source: arXiv

Progress summary

Refreshed
Open · progress reported

A new preprint proves quadratic degree growth for an infinite subfamily of these recurrences, but the general conjecture remains open.

The problem asks whether all three-term Gale–Robinson recurrences have quadratic degree growth, possibly with periodic corrections. The general assertion is not settled.

October 2026 infinite-family result

Melih Emin Can’s unrefereed preprint reduces degree growth to tropical recurrences, gives formulas for broad quadratic Laurent recurrences, and proves the proposed formula for the infinite family $p=1, q=2$. This establishes substantial progress but does not resolve the conjecture for general three-term parameters.

Current status (as of October 2026): quadratic growth is claimed for the infinite $p=1, q=2$ family, while the general three-term conjecture remains open and the new result is unverified.

Sources

Solutions 0

No solutions have been posted yet.