CNN feature activations follow long-tailed Weibull-like distributions with increasing tail dependence by depth rather than Gaussian, indicating a Matthew process that concentrates signal in tails.
Diffusion models: A comprehensive survey of methods and applications
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
Topological features of interaction graphs determine whether locally interacting systems on planar graphs can sustain ordered phases, with implications for biological self-organization versus limitations in autoregressive models.
citing papers explorer
-
Why CNN Features Are not Gaussian: A Statistical Anatomy of Deep Representations
CNN feature activations follow long-tailed Weibull-like distributions with increasing tail dependence by depth rather than Gaussian, indicating a Matthew process that concentrates signal in tails.
-
Topological constraints on self-organisation in locally interacting systems
Topological features of interaction graphs determine whether locally interacting systems on planar graphs can sustain ordered phases, with implications for biological self-organization versus limitations in autoregressive models.