Implicit Generation and Generalization with Energy Based Models

By Yilun Du, Igor Mordatch. 2018

Subfields: Generative Models, Probabilistic & Bayesian Methods

How researchers see this paper

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

In Generative Models

In Probabilistic & Bayesian Methods

Citations in the corpus

Residual Energy-Based Models for Text Generation (2020, methodology): “In maximum likelihood training negatives are generated from the model, but in text application we cannot use gradient-based MCMC methods (Teh et al., 2003; Du & Mordatch, 2019) and Gibbs sampling (Welling et al., 2005) is too slow to be practical.”
Your Classifier is Secretly an Energy Based Model and You Should Treat it Like One (2019, methodology): “Regularizers may be added (Du & Mordatch, 2019) to increase stability but it is not clear what effect they have on the final model.”
Flow Contrastive Estimation of Energy-Based Models (2019, background): “[55, 9] scale up sampling-based methods to large image datasets with white noise as the starting point of sampling.”
Energy-Inspired Models: Learning with Sampler-Induced Distributions (2019, background): “Empirically, previous work has found that convolutional architectures that score images ( i.e. , map x to a real number) tend to have strong inductive biases that match natural data [22].”
Generative Modeling by Estimating Gradients of the Data Distribution (2019, methodology): “(4), but it can often be ignored in practice [9, 12, 39].”
Learning Non-Convergent Non-Persistent Short-Run MCMC Toward Energy-Based Model (2019, background): “Recent work that parametrizes the energy function by modern convolutional neural networks (ConvNets) [28, 26] suggests that the “analysis by synthesis” process can indeed generate highly realistic images [48, 11, 21, 10].”
On the Anatomy of MCMC-based Maximum Likelihood Learning of Energy-Based Models (2019, background): “The authors of [22, 6] store a large set of persistent images and sample batches of images from the set for each parameter update.”
D2C: Diffusion-Decoding Models for Few-Shot Conditional Generation (2021, background): “Latent variable generative models Most deep generative models explicitly define a latent representation, except for some energy-based models [41, 31] and autoregressive models [94, 93, 10].”
Diffusion Models Beat GANs on Image Synthesis (2021, background): “Du and Mordatch [15] further improve upon this approach, obtaining high quality images.”
On Buggy Resizing Libraries and Surprising Subtleties in FID Calculation (2021, background): “A wide range of image and video synthesis applications [45, 54, 67, 83] have been enabled, as a result of tremendous progress in deep generative models such as GANs [9, 28, 34, 37, 59], VAEs [41, 55, 61], autoregressive models [52], flow-based models [18, 40], and energy-based models [21, 50, 63].”
Implicit Behavioral Cloning (2021, methodology): “Constructive Approximation, pages 1–39, 2021.339 [34] I. Mordatch.”
Learning Energy-Based Models by Diffusion Recovery Likelihood (2020, methodology): “The approach is adopted by [35] and [5].”
Energy-based Out-of-distribution Detection (2020, neutral): “In Table 3, we also compare with state-of-the-art hybrid models that incorporated generative modeling [8, 11, 18].”
Real or Fake? Learning to Discriminate Machine from Human Generated Text (2019, neutral): “In settings with continuous variables, researchers have suggested MCMC (Teh et al., 2003) or Langevin dynamics (Du & Mordatch, 2019).”
Energy-based models for atomic-resolution protein conformations (2020, methodology): “This is motivated by their simplicity and scalability, as well as their compelling results in other domains, such as image generation (Du & Mordatch, 2019).”
PixelVAE++: Improved PixelVAE with Discrete Prior (2019, background): “…of success by latent variable models such as variational autoencoder (VAE) [19, 20, 21, 4], ﬂow-based models [22, 23, 24, 5], energy-based model [25], autoregressive models such as PixelCNN [26, 1, 27, 6, 28, 29], and lately, by the combination of latent-variable and autoregressive models,…”