Infinite-dimensional stochastic differential equations arising from Airy random point fields (Q6956182)

The \(Airy_{\beta}\) random point fields (\(\beta = 1,2,4 \)) are random point fields, corresponding to soft-edge scaling limits of eigenvalues of Gaussian random matrices (e.g. GOE, GUE, GSE). The article under review studies infinite-dimensional stochastic differential equations arising from them. It starts by constructing a reversible unlabelled diffusion. It then derives the infinite-dimensional stochastic differential equations based on this diffusion construction and studies them using various stochastic tools, including Plancherel-Rotach and strong asymptotics. Palm probability measures are defined for the purpose of studying the problem, and estimates of correlation functions are derived. The authors prove the existence and pathwise uniqueness of strong solutions for the SDE. They observe and study an interesting correspondence to the dynamics of the \(N\)-particle systems, putting the analysis in the framework of an interacting particle system with time-dependent dynamics. All the results are derived for the soft-edge limit which is defined accordingly. The special case of \(\beta = 2\) is studied, correspondingly to \(\beta = 1\) and \(\beta = 4\), as the predominant cases in the study of random matrix ensembles. In long (more than 50-pages) appendices, supporting theory of Airy functions and ensembles, asymptotics for oscillator functions and detailed proof derivations are included.\N\NThe authors conjecture that their approach is valid also for other soft-edge scaling limits of the stochastic dynamics arising in random matrix theory. Interesting additions could go in the direction of studying the bulk properties, general beta ensembles (e.g., the case of \(\beta = 6\)) and in a non-asymptotic setting of concentration inequalities, frequently used to study various random matrix objects.