Abstract:We establish the universality of the singular numbers in random matrix products over $\mathrm{GL}_n(\mathbb{Q}_p)$ as the number of products approaches infinity, with a fixed $n\ge 1$. We demonstrate that, under a broad class of distributions, which we term as ``split", the asymptotics of matrix products align with the sum of the singular numbers of the matrix corners. Specifically, when matrices are independent and identically distributed, we derive the strong law of large numbers and the central limit theorem. Our approach is inspired by Van Peski's work (<a class="link-https" data-arxiv-id="2011.09356" href="https://arxiv.org/abs/2011.09356">arXiv:2011.09356</a>), which examines products of $n\times n$ corners of Haar-distributed elements $\mathrm{GL}_N(\mathbb{Z}_p)$. We extend the method so that the criterion now works as long as the measures are left- and right-invariant under the multiplication of $\mathrm{GL}_n(\mathbb{Z}_p)$. Building on this approach for $\mathrm{GL}_n$, we further demonstrate similar universality for $\mathrm{GSp}_{2n}$ and discuss potential extensions to general split reductive groups.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is the distribution characteristics of the singular numbers of random matrix products over the \(p\)-adic number field \(\mathbb{Q}_p\) when the number of products tends to infinity. Specifically, the author aims to prove that under the generalized "split" distribution, the asymptotic behavior of these singular numbers is the same as that of the sum of the singular numbers of matrix corners, and in the case of independent and identically distributed (i.i.d.), the strong law of large numbers and the central limit theorem can be derived.
### Main research questions:
1. **Universality of singular numbers**: Investigate whether the asymptotic behavior of singular numbers in random matrix products over \(\text{GL}_n(\mathbb{Q}_p)\) is universal, that is, does not depend on the specific initial distribution.
2. **Strong law and central limit theorem**: In the case of independent and identically distributed, prove the strong law and the central limit theorem of singular numbers.
3. **Extension under split distribution**: Explore whether the above results still hold under a broader "split" distribution and discuss their applications in other split reductive groups (such as \(\text{GSp}_{2n}\)).
### Key concepts:
- **Singular Numbers**: For a non - singular \(p\)-adic matrix \(A\), its singular numbers are defined as the diagonal elements \((p^{\lambda_1}, p^{\lambda_2},\ldots, p^{\lambda_n})\) after diagonalization, where \(\lambda_i\) are integers.
- **Split Distribution**: A class of distributions that are invariant under left and right multiplication by \(\text{GL}_n(\mathbb{Z}_p)\).
- **Matrix Corners**: Sub - matrices extracted from a matrix, especially the sub - matrix consisting of the last row and all columns.
### Research methods:
1. **Heuristic method**: By assuming the distribution invariance of matrix products, the problem is transformed into analyzing the singular numbers of matrix corners.
2. **Rigorous proof**: Use methods of linear algebra and probability theory to rigorously prove the effectiveness of the above heuristic method.
3. **Generalization to split reductive groups**: Generalize the results to other split reductive groups, such as \(\text{GSp}_{2n}\), and prove similar strong laws and central limit theorems.
### Main conclusions:
- **Strong law**: In the case of independent and identically distributed, the average value of the singular numbers of matrix products converges to the expected value of the singular numbers of matrix corners.
- **Central limit theorem**: In the case of independent and identically distributed, the standardized deviation of the singular numbers of matrix products approaches a normal distribution.
- **Universality**: Under a broader "split" distribution, the above results still hold.
### Significance:
This research not only deepens the understanding of random matrix products over the \(p\)-adic number field but also provides a theoretical basis for similar problems on other split reductive groups. These results have important application values in fields such as statistical physics and theoretical computer science.