sampling
TL;DR: This paper introduces Consistency Models, a new family of models based on diffusion models that enables 1-step generation.
Consistency models are a new family of models based on continuous-time diffusion models [Song et al. ICLR 2021], [Karras el al. NeurIPS 2022] that achieve 1-step generation.
Continuous-time diffusion models [Song et al. ICLR 2021] are formulated with the stochastic differential equation (SDE): \[ \text{d}\mathbf{x}_t = \underbrace{\boldsymbol{\mu}(\mathbf{x}_t, t) \ \text{d}t}_{\text{Deterministic Term}} + \underbrace{\sigma(t) \ \text{d}\mathbf{w}_t}_{\text{Stochastic Term}}, \label{eq:ode} \tag{Eq. 1} \] where:
\[ \mathcal{L}_{CD}^N(\boldsymbol{\theta}, \boldsymbol{\theta}^{-}; \phi) = \mathbb{E} [\lambda(t_n) d(\boldsymbol{f_\theta}(\mathbf{x}_{t_{n*1}}, t_{n+1}), \boldsymbol{f_{\theta{-}}}(\mathbf{\hat{x}^\phi}_{t_n}, t_n))] \label{eq:loss} \tag{Eq. X} \] rember that the expectation \( \mathbb{E} \) is like a mean but without samples, so it is a theoretical mean of the distribution.
© Copyright © All rights reserved | Carlos Hernández Oliván | Colorlib