Work

Spectral Independence Beyond Total Influence on Trees and Related Graphs

Year: 2025
Type: conference-paper
Abstract: We study how to establish spectral independence, a key concept in sampling, without relying on total influence bounds, by applying an approximate inverse of the influence matrix. Our method gives constant upper bounds on spectral independence for two well-studied Gibbs distributions known to have unbounded total influences:
Language: English
Open Access status: green
OpenAlex ID: W4406142467
FWCI: 3.266
Cites: 53
Cited by: 5
Related to: 9