Abstract:For a graph $G$ and a hereditary property $\mathcal{P}$, let $\text{ex}(G,\mathcal{P})$ denote the maximum number of edges of a subgraph of $G$ that belongs to $\mathcal{P}$. We prove that for every non-trivial hereditary property $\mathcal{P}$ such that $L \notin \mathcal{P}$ for some bipartite graph $L$ and for every fixed $p \in (0,1)$ we have \[\text{ex}(G(n,p),\mathcal{P}) \le n^{2-\varepsilon}\] with high probability, for some constant $\varepsilon = \varepsilon(\mathcal{P})>0$. This answers a question of Alon, Krivelevich and Samotij.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about estimating the maximum number of edges of subgraphs with specific hereditary properties in random graphs. Specifically, the authors study the upper bound of the number of edges of the largest subgraph belonging to \(P\) in the Erdős–Rényi random graph \(G(n, p)\) for a given hereditary property \(P\).
### Background and Problem Description
1. **Hereditary Property \(P\)**: A hereditary property refers to a set of graphs in which every induced subgraph of each graph in the set must also belong to this set. For example, if a graph \(G\in P\), then any induced subgraph \(G[S]\) of \(G\) must also be in \(P\).
2. **Maximum Number of Edges \(ex(G, P)\)**: For a graph \(G\) and a hereditary property \(P\), \(ex(G, P)\) represents the maximum possible number of edges in the subgraphs of \(G\) that belong to \(P\).
3. **Core of the Problem**: The paper focuses on how to estimate the upper bound of \(ex(G(n, p), P)\) when the hereditary property \(P\) lacks a certain bipartite graph \(L\), that is, \(L\notin P\). In particular, Alon, Krivelevich and Samotij proposed a question: Does there exist a constant \(\varepsilon=\varepsilon(P)> 0\) such that when \(P\) lacks a certain bipartite graph \(L\),
\[
ex(G(n, p), P)\leq n^{2-\varepsilon}
\]
holds with high probability?
### Main Contributions of the Paper
The paper answers the above - mentioned question by proving the following theorem:
**Theorem 1.1**: Let \(P\) be a hereditary graph property and there exists a bipartite graph \(L\notin P\). Then for any fixed \(p\in(0, 1)\), with high probability
\[
ex(G(n, p), P)\leq n^{2-\varepsilon}
\]
where \(\varepsilon = \varepsilon(L)>0\) is a constant.
### Method Overview
To prove this theorem, the authors combine Kővári–Sós–Turán - type arguments and recent results (Bourneuf et al.). Specific steps include:
- **Lemma 2.2**: Prove the probability that the number of independent sets in the random graph \(G(n, p)\) satisfies certain conditions.
- **Theorem 2.1**: Deal with the special case of the complete bipartite graph \(K_{s,t}\), and prove that \(ex_I(G(n, p), K_{s,t})=O(n^{2 - 1/s}(\log n)^{1/s})\).
- **Lemma 3.1**: A key lemma for dealing with the general case of the bipartite graph \(L\), and finally prove Theorem 1.1 by induction.
Through these steps, the paper successfully answers the question proposed by Alon, Krivelevich and Samotij and provides a rigorous mathematical proof.