Exponential Tilting of Generative Models: Improving Sample Quality by Training and Sampling from Latent Energy
arXiv:2006.08100
Abstract
In this paper, we present a general method that can improve the sample quality of pre-trained likelihood based generative models. Our method constructs an energy function on the latent variable space that yields an energy function on samples produced by the pre-trained generative model. The energy based model is efficiently trained by maximizing the data likelihood, and after training, new samples in the latent space are generated from the energy based model and passed through the generator to producing samples in observation space. We show that using our proposed method, we can greatly improve the sample quality of popular likelihood based generative models, such as normalizing flows and VAEs, with very little computational overhead.
arXiv comment15 pages
References in corpus (13)
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Neural Discrete Representation Learning
- Stochastic Backpropagation and Approximate Inference in Deep Generative Models
- Improved Techniques for Training GANs
- Variational Graph Auto-Encoders
- Invertible Residual Networks
- Implicit Generation and Generalization in Energy-Based Models
- Diagnosing and Enhancing VAE Models
- Your Classifier is Secretly an Energy Based Model and You Should Treat it Like One
- Augmented Normalizing Flows: Bridging the Gap Between Generative Flows and Latent Variable Models
- Generative Latent Flow
- Metropolis-Hastings Generative Adversarial Networks