SearcharxivSearch

arXiv subjects

Dawson Thomas

Publications and source records attributed to Dawson Thomas.

2 recordsLinked to original sources

Wet Hair: Global Symmetries in Entanglement Islands

A central conjecture in quantum gravity is the non-existence of global symmetries. As a fully unitary theory, there is no information loss in a UV complete quantum gravity theory. We see both these concepts reflected in the AdS/CFT correspondence, which tells us that dynamical processes in AdS are fully captured by a manifestly unitary CFT with no information loss. Furthermore, global symmetries of the CFT are dual to gauge symmetries in the AdS, which implies no global symmetry in the AdS. In this work, we provide concrete evidence for the connection between the non-existence of global symmetries and the absence of information loss in quantum gravity. We study the $\textit{island setups}$ in which a gravitational AdS is coupled with a nongravitational bath on its boundary. In such theories, the information in the AdS can be lost to the bath. We provide concrete examples with global symmetries in the island setup, from both the bottom-up and the top-down perspectives. We argue that these global symmetries are consistent due to $\textit{entanglement islands}$, in which holography is realized in a novel fashion. The global symmetries we construct are all mixed with spontaneously broken gauge symmetries. We will show that this fact has two implications: $\textbf{1)}$ The black hole hair is detectable in the bath (``wet hair"); $\textbf{2)}$ a resolution of a puzzle proposed by Harlow and Shaghoulian.

hep-th

Diffusion Curvature for Estimating Local Curvature in High Dimensional Data

We introduce a new intrinsic measure of local curvature on point-cloud data called diffusion curvature. Our measure uses the framework of diffusion maps, including the data diffusion operator, to structure point cloud data and define local curvature based on the laziness of a random walk starting at a point or region of the data. We show that this laziness directly relates to volume comparison results from Riemannian geometry. We then extend this scalar curvature notion to an entire quadratic form using neural network estimations based on the diffusion map of point-cloud data. We show applications of both estimations on toy data, single-cell data, and on estimating local Hessian matrices of neural network loss landscapes.

cs.LG