Abstract:Let $Z_1,\, Z_2,\dots$ be independent and identically distributed complex random variables with common distribution $\mu$ and set $$ P_n(z) := (z - Z_1)\cdots (z - Z_n)\,. $$ Recently, Angst, Malicet and Poly proved that the critical points of $P_n$ converge in an almost-sure sense to the measure $\mu$ as $n$ tends to infinity, thereby confirming a conjecture of Cheung-Ng-Yam and Kabluchko. In this short note, we prove for any fixed $k\in \mathbb{N}$, the empirical measure of zeros of the $k$th derivative of $P_n$ converges to $\mu$ in the almost sure sense, as conjectured by Angst-Malicet-Poly.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the empirical measure of the zeros of the derivatives of random polynomials converging to a given probability measure \(\mu\) in the almost - sure sense. Specifically, for any fixed \(k\in\mathbb{N}\), the author proves that the empirical measure \(\nu_n^{(k)}\) of the zeros of the \(k\)-th derivative of the random polynomial \(P_n(z)\) converges almost surely to \(\mu\).
### Background and Motivation
1. **Random Polynomial Model**:
- Let \(\mu\) be a probability measure on the complex plane, and \((Z_j)_{j\geq1}\) be independent and identically distributed complex random variables with common distribution \(\mu\).
- Define the sequence of random polynomials \((P_n)_{n\geq1}\) as:
\[
P_n(z):=(z - Z_1)(z - Z_2)\cdots(z - Z_n).
\]
2. **Existing Results**:
- Pemantle and Rivin conjectured that the critical points of \(P_n\) (i.e., the zeros of \(P_n'\)) are close to the roots of \(P_n\).
- Kabluchko proved that when \(k = 1\), \(\nu_n^{(1)}\) converges to \(\mu\) with probability 1.
- Byun, Lee and Reddy proved that for a fixed \(k\in\mathbb{N}\), \(\nu_n^{(k)}\) converges to \(\mu\) with probability 1.
3. **New Conjecture**:
- Angst, Malicet and Poly conjectured that for a fixed \(k\in\mathbb{N}\), \(\nu_n^{(k)}\) should converge almost surely to \(\mu\).
### Main Contributions
1. **Main Theorem**:
- The author proves that for any fixed \(k\in\mathbb{N}\), \(\nu_n^{(k)}\) converges almost surely to \(\mu\).
- Specifically:
\[
\text{Almost surely, for each fixed } k\in\mathbb{N},\text{ the sequence } \nu_n^{(k)} \text{ converges to } \mu \text{ as } n\rightarrow\infty.
\]
2. **Proof Method**:
- The author draws on the proof method of Angst, Malicet and Poly for the case \(k = 1\) and introduces a new decoupling technique to deal with high - dimensional multi - variable anti - concentration estimation problems.
- By using Moebius transformation and Jensen's inequality, the author transforms the problem into a high - dimensional linear anti - concentration estimation problem.
- By applying the Borel - Cantelli lemma, the author proves the required almost - sure convergence.
### Technical Details
1. **Anti - Concentration Estimation**:
- The author uses the decoupling lemma introduced by Costello, Tao and Vu to deal with high - dimensional multi - variable anti - concentration estimation problems.
- By transforming the multi - linear problem into a linear problem, the author can obtain the required probability decay estimate.
2. **Key Lemma**:
- Lemma 1.3: Assume that \(\mu\) has no finite support. For \(L = 2^k+2k\) and all pairwise distinct complex numbers \((z_1,\ldots,z_L)\), we have
\[
P(|S_n(z_j)|\leq1\text{ for all }j\in[L])\leq\frac{C}{n^2}
\]
where \(C>0\) depends on \(k\) and \(\mu\).
### Conclusion