The statistical thermodynamics of generative diffusion models: Phase transitions, symmetry breaking and critical instability

arXiv:2310.17467v2 Announce Type: replace-cross
Abstract: Generative diffusion models have achieved spectacular performance in many areas of generative modeling. While the fundamental ideas behind these models come from non-equilibrium physics, variational inference and stochastic calculus, in this paper we show that many aspects of these models can be understood using the tools of equilibrium statistical mechanics. Using this reformulation, we show that generative diffusion models undergo second-order phase transitions corresponding to symmetry breaking phenomena. We show that these phase-transitions are always …

abstract arxiv breaking calculus cs.lg diffusion diffusion models equilibrium generative generative modeling ideas inference modeling paper performance physics show statistical stat.ml stochastic symmetry transitions type