Events

Date: 16 January 2026   Time: 11:00 - 12:00

Location: 114 G.O. Jones Building

Title: Holography and optimal transport

Abstract: Optimal transport and Wasserstein distance are often used in machine learning, in particular in diffusion models in which
Fokker-Planck equation of the probability flow is optimized. We discuss how the idea of emergent spacetime in holography can
fit the notion of optimal transport. We employ the simplest example of a single quantum harmonic oscillator and demonstrate
that the Wasserstein distance of the optimal transport between states gives rise to a holographic geometry. Furthermore, we can
identify the Wasserstein distance as a generalized Krylov complexity.
(This work is in collaboration with Norihiro Tanahashi, work in progress.)