Who killed the directed model?

33 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modifications (2017, methodology): “Our approach follows earlier work using continuous mixture models (Domke et al., 2008; Theis et al., 2012; Uria et al., 2013; Theis & Bethge, 2015), but avoids allocating probability mass to values outside the valid range of [0 , 255] by explicitly modeling the rounding of ν to x .”
Generative Image Modeling Using Spatial LSTMs (2015, methodology): “n ordering and specify the distribution of each pixel conditioned on its parent pixels. Several parametrizations have been suggested for the conditional distributions in the context of natural images [5, 15, 41, 44, 45]. We here review and extend the work of Theis et al. [41] who proposed to use mix”
Fast Lossless Neural Compression with Integer-Only Discrete Flows (2022, neutral): “…et al. (2020b); Lu et al. (2021), NFs), variational auto-encoders (Kingma & Welling (2013); Rezende et al. (2014), VAEs), auto-regressive models (Domke et al. (2008); Larochelle & Murray (2011); Salimans et al. (2017), ARMs), and diffusion models (Sohl-Dickstein et al. (2015); Ho et al. (2020);…”
Visual Commonsense R-CNN (2020, unknown): “Un-/Self-supervised Visual Feature Learning [14, 63, 43, 29, 77].”
From learning models of natural image patches to whole image restoration (2011, methodology): “[9, 1, 14]) but they typically constrain the means and covariance matrices during learning.”