ProbXiv
Explorersign in
Problem archiveProblem record

Statement

Let G=(V,E,s)G=(V,E,s) be an undirected sandpile graph. For U⊆VU\subseteq V, let G∣UG|_U be the subgraph induced by UU. A connected kk-partition of GG is a partition Π=⨆i=1kVi\Pi=\bigsqcup_{i=1}^k V_i such that each G∣ViG|_{V_i} is connected. Let GΠG_\Pi be the graph whose vertices are V1,…,VkV_1,\dots,V_k, with edge weights

wt⁡(Vi,Vj)=#{e∈E:one endpoint of e is in Vi and the other is in Vj}.\operatorname{wt}(V_i,V_j)=\#\{e\in E:\text{one endpoint of }e\text{ is in }V_i\text{ and the other is in }V_j\}.

Treat GΠG_\Pi as a sandpile graph with sink vertex ViV_i for the unique ii such that s∈Vis\in V_i, and let Pk\mathcal{P}_k denote the set of connected kk-partitions of GG. The sandpile Betti-number conjecture.

βk=∑Π∈Pk+1#{c:c a minimal recurrent configuration on GΠ}.\beta_k=\sum_{\Pi\in\mathcal{P}_{k+1}}\#\{c:\text{$c$ a minimal recurrent configuration on }G_\Pi\}.

This conjecture identifies the Betti numbers of the relevant sandpile-related algebra with counts of minimal recurrent configurations on partition graphs. It appears as Corollary 3.29 in work of Wilmes and has been verified using Sage for all undirected, unweighted graphs with fewer than 77 vertices; the general assertion remains open.

Record

Comments

Nothing has been published against this problem yet, and nothing has been said about it.

Sign in with an institutional address to take part in the discussion. Reading every thread stays open to everyone.

Sign in

Solve with an agent

Open the statement in a chat, with the problem and the ground rules already written into the prompt.

This opens a third-party site. Nothing is posted back to ProbXiv and nothing you write there is recorded here — what a model gives you is an attempt, which a person still has to check.