notesum.ai

Published at November 4

Unsupervised detection of semantic correlations in big data

cs.LG

cs.AI

physics.comp-ph

Released Date: November 4, 2024

Authors: Santiago Acevedo¹, Alex Rodriguez², Alessandro Laio³

Aff.: ¹International School for Advanced Studies (SISSA), Via Bonomea 265, 34136 Trieste, Italy; ²Dipartimento di Matematica e Geoscienze, Università degli Studi di Trieste, via Alfonso Valerio 12/1, 34127 Trieste, Italy; The Abdus Salam International Centre for Theoretical Physics (ICTP), Strada Costiera 11, 34014 Trieste, Italy; ³International School for Advanced Studies (SISSA), Via Bonomea 265, 34136 Trieste, Italy; The Abdus Salam International Centre for Theoretical Physics (ICTP), Strada Costiera 11, 34014 Trieste, Italy

Arxiv: http://arxiv.org/abs/2411.02126v1

Number of blocks $(B)$	1	10	100
Number of bits $(L)$	$10^{4}$	$10^{5}$	$10^{6}$
BID	8047	80497	804624
BID per bit	0.8047	0.8050	0.8046
log(KL divergence)	-9.2	-8.2	-7.1