What problem does this paper attempt to address?

The main problem that this paper attempts to solve is about the non - saturation probability in the d - process of random graphs. Specifically, the paper focuses on the probability that the final graph is not saturated in the d - process, that is, the probability that there are vertices with degrees less than d in the final graph. By introducing a ball - box model closely related to the d - process, the paper provides a new analytical method, which enables this probability to be determined more accurately. ### Background of the Paper In the d - process, starting from an empty graph, an edge is added each time, and this edge is uniformly and randomly selected from all pairs of vertices whose current degrees are less than d. If there is at most one vertex with degree d - 1 in the final graph and all other vertices have degree d, then this process is called saturated. Otherwise, it is called non - saturated. ### Research Questions The main research questions of the paper are: - **Determine the probability of non - saturation in the d - process**: Specifically, when there are vertices with degrees less than d in the final graph, what is the probability of this non - saturated situation? - **Improve the estimation of degree distribution**: How can the degree distribution at each moment in the entire d - process be described more accurately? ### Main Contributions 1. **New analytical method**: The paper introduces a ball - box model related to the d - process. Through this model, the degree distribution in the d - process can be analyzed more simply. 2. **More accurate results**: Using the new method, the paper obtains more accurate results regarding the non - saturation probability, especially providing a more detailed answer to the question of whether the final graph of the d - process is saturated. 3. **Extension of known results**: The paper not only re - proves some known results but also obtains some new and more refined results, especially in the later stages of the d - process. ### Specific Results - **Theorem 1.2**: Gives the asymptotic expression for the probability of non - saturation in the d - process: \[ P(F)\sim\begin{cases} \frac{d - 1}{\log n}&\text{when }dn\text{ is even}\\ \frac{(d - 1)(d - 2)}{\log^{2}n}&\text{when }dn\text{ is odd} \end{cases} \] This result confirms a previous conjecture and provides a more accurate answer. - **Theorem 1.4**: For each \(j < d\), in the early stage of the d - process, the concentration property of the vertex degree \(D_j(s)\), that is, \(D_j(s)\) changes within a relatively narrow window with high probability. - **Theorem 1.5 and 1.6**: Further study the later stages of the d - process, especially when approaching the maximum possible number of edges, the expected value and distribution of the vertex degrees. ### Methodology The paper links the d - process with a simpler random allocation model by introducing the ball - box model. This method not only simplifies the analysis but also provides a more accurate estimation of the degree distribution, thereby enabling a better understanding of the properties of the d - process. ### Conclusion By introducing a new analytical method, the paper successfully solves the non - saturation probability problem in the d - process and provides a more accurate estimation of the degree distribution. These results not only verify previous conjectures but also provide new tools and perspectives for future related research.

Sharper analysis of the random graph $d$-process via a balls-in-bins model

Local limit of the random degree constrained process

Balls into bins via local search: cover time and maximum load

Annealed invariance principle for random walks on random graphs generated by point processes in $\mathbb{R}^d$

The stable graph: The metric space scaling limit of a critical random graph with i.i.d. power-law degrees

Isomorphisms between dense random graphs

A Generalization of Multiple Choice Balls-into-Bins: Tight Bounds

Majority Model on Random Regular Graphs

Random graphs from a weighted minor-closed class

The Power of Two Choices in Graphical Allocation

Evolution of locally dependent random graphs

Hitting times in the binomial random graph

Large induced subgraphs of random graphs with given degree sequences

The Low-Degree Hardness of Finding Large Independent Sets in Sparse Random Hypergraphs

Using random graphs to sample repulsive Gibbs point processes with arbitrary-range potentials

$d$-connectivity of the random graph with restricted budget

The Spectral Distribution of Random Graphs with Given Degree Sequences

Chains-into-Bins Processes

Random Growth Networks with Exponential Degree Distribution

The computational hardness of counting in two-spin models on d-regular graphs

Sandwiching Random Geometric Graphs and Erdos-Renyi with Applications: Sharp Thresholds, Robust Testing, and Enumeration