אריה קנטורוביץ

אקדמי בכיר

On the tensorization of the variational distance

If one seeks to estimate the total variation between two product measures (formula Presented) in terms of their marginal TV sequence (Math Presented) then trivial upper and lower bounds are provided by (formula Presented) We improve the lower bound to (formula Presented) TV, there by reducing the gap between the upper and lower bounds from (formula Presented) Furthermore, we show that any estimate on (formula Presented) expressed in terms of δ must necessarily exhibit a gap of (formula Presented) between the upper and lower bounds in the worst case, establishing a sense in which our estimate is optimal. Finally, we identify a natural class of distributions for which (formula Presented) approximates the TV distance up to absolute multiplicative constants.

שפת פרסום אנגלית
כתב עת Electronic Communications in Probability
כרך 30
סטטוס פרסום פורסם - 01.01.2025
מספר מאמר 32

Keywords

Euclidean distance
tensorization
total variation

ASJC Scopus subject areas

Statistics and Probability
Statistics, Probability and Uncertainty
גישה למסמך
10.1214/25-ECP680
קבצים וקישורים אחרים
Link to publication in Scopus