Chaotic D1-D5 Black Hole Dynamics through Networks
This work studies dynamics controlling the transition between different microstates of two charge D1-D5 black holes by network methods, in which microstates of the system are defined as network nodes, while transitions between them are defined as edges. It is found that the eigenspectrum of this network's Laplacian matrix, which is identified with Hamiltonians of the microstate system, has completely the same Nearest-Neighbor Spacing Distribution as that of general Gaussian Orthogonal Ensemble of Random Matrices. According to the BGS, i.e. Bohigas, Giannoni and Schmit conjecture, this forms evidence for chaotic features of the D1-D5 microstate dynamics. This evidence is further strengthened by observations that inverse of the first/minimal nonzero eigenvalue of the Laplacian matrix is proportional to logarithms of the microstate number of the system. By Sekino and Susskind, this means that dynamics of the D1-D5 black hole microstates are not only chaotic, but also the fastest scrambler in nature.