Abstract:The stochastic sandpile model (SSM) is a generalisation of the standard Abelian sandpile model (ASM), in which topplings of unstable vertices are made random. When unstable, a vertex sends one grain to each of its neighbours independently with probability $p \in (0,1)$. We study the SSM on complete graphs. Our main result is a description of the recurrent states of the model. We show that these are given by convex sums of recurrent states for the ASM. This allows us to recover a well-known result: that the number of integer lattice points in the $n$-dimensional permutation polytope is equal to the number of labeled spanning forests on $n$ vertices. We also provide a stochastic version of Dhar's burning algorithm to check if a given (stable) state is recurrent or not, which runs in linear time. Finally, we study a family of so-called "partial" SSMs, in which some vertices topple randomly, while others topple deterministically (as in the ASM, sending one grain to all neighbours). We show that this distinction is meaningful, yielding sets of recurrent states that are in general different from those of both the ASM and SSM. We also show that to get all recurrent states of the SSM, we can allow up to two vertices to topple deterministically.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to study the recurrent states of the Stochastic Sandpile Model (SSM) on a complete graph. Specifically, the author aims to provide a combinatorial description of these recurrent states and show that these states can be represented by integer lattice points within a convex polytope. In addition, the paper also proposes a stochastic version of the Dhar burning algorithm, which is used to check whether a given stable state is a recurrent state, and this algorithm has a linear time complexity. Finally, the paper explores the partial Stochastic Sandpile Model (partial SSM), in which some vertices collapse randomly while others collapse deterministically, and shows that the set of recurrent states of these models is different from that of the standard Abelian Sandpile Model (ASM) and the complete Stochastic Sandpile Model (SSM).
### Main contributions of the paper
1. **Combinatorial description of recurrent states**:
- The paper proves that the recurrent states of SSM on a complete graph can be given by convex combinations of the recurrent states of ASM. This provides a new perspective for understanding the recurrent states of SSM.
- Specifically, these recurrent states correspond to the number of integer lattice points within the n - dimensional permutation polytope, and this result matches the known number of recurrent states of ASM.
2. **Stochastic burning algorithm**:
- A stochastic version of the Dhar burning algorithm is proposed, which can determine whether a given stable state is a recurrent state in linear time.
3. **Partial stochastic sandpile model**:
- The partial stochastic sandpile model, in which some vertices collapse randomly while others collapse deterministically, is studied.
- It is proved that the set of recurrent states of these models is different from that of ASM and SSM.
- It is further proved that when at most two vertices are allowed to collapse deterministically, all the recurrent states of SSM can be obtained.
### Mathematical background
- **Sandpile model**:
- **Abelian sandpile model (ASM)**: On a graph, a certain number of sand grains are assigned to each vertex. When the number of sand grains at a vertex exceeds its degree, the vertex becomes unstable and will evenly distribute its sand grains to its neighbors.
- **Stochastic sandpile model (SSM)**: Randomness is introduced on the basis of ASM, that is, when a vertex is unstable, it will randomly select one or more vertices among its neighbors to send sand grains with probability \( p \).
- **Recurrent states**:
- Recurrent states refer to the states that appear infinitely many times in the long - running model.
- For ASM, recurrent states can be determined in multiple ways, such as through the Dhar burning algorithm or compatible acyclic oriented graphs.
- **Combinatorial objects**:
- **Parking functions**: On a complete graph, there is a bijection between the recurrent states of ASM and parking functions.
- **Permutation polytope**: The number of integer lattice points within the n - dimensional permutation polytope is equal to the number of labeled spanning forests of n vertices.
### Conclusion
The paper extends the understanding of ASM by providing a combinatorial description of the recurrent states of SSM on a complete graph, and provides new tools and methods for the study of stochastic sandpile models. These results are not only of great theoretical significance, but also provide strong support for the analysis of sandpile models in practical applications.