Diffusion and flow processes forget dependencies to define valid copulas then learn to remember them for density estimation and sampling, outperforming prior copula methods on complex datasets.
Towards a universal representation of statistical dependence
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
stat.ML 2verdicts
UNVERDICTED 2representative citing papers
Deep neural networks trained to classify simulated samples under null and alternative hypotheses produce a test statistic that outperforms nineteen competing methods for independence testing across varied dependence structures.
citing papers explorer
-
Deep-testing: the case of dependence detection
Deep neural networks trained to classify simulated samples under null and alternative hypotheses produce a test statistic that outperforms nineteen competing methods for independence testing across varied dependence structures.