Large Deviations for the Largest Eigenvalue of Gaussian Kronecker Random Matrices

par Jana Reker

jeudi 8 janv. 2026, 14:00 → 15:00 Europe/Paris

435 (ENS de Lyon)

We consider Gaussian Kronecker random matrices of the form $X^{(N)}:=\sum_{j=1}^k A_j\otimes W_j+A_0\otimes Id$, where $A_0, ..., A_k$ are real symmetric (resp. complex Hermitian) deterministic $L\times L$ matrices, $W_1, ..., W_k$ are sampled independently from the GOE (resp. GUE) of size $N\times N$, and Id denotes identity. In this setting, we show a large deviations principle for the largest eigenvalue in the regime where the dimension of the Gaussian matrices goes to infinity. The talk is based on joint work with A. Guionnet and J. Husson.