Learning Latent Space Energy-Based Prior Model

NeurIPS 2020

The generator model assumes that the observed example is generated by a low-dimensional latent vector via a top-down network, and the latent vector follows a simple and known prior distribution, such as uniform or Gaussian white noise distribution. While we can learn an expressive top-down network to map the prior distribution to the data distribution, we can also learn an expressive prior model instead of assuming a given prior distribution. This follows the philosophy of empirical Bayes where the prior model is learned from the observed data. We propose to learn an energy-based prior model for the latent vector, where the energy function is parametrized by a very simple multi-layer perceptron. Due to the low-dimensionality of the latent space, learning a latent space energy-based prior model proves to be both feasible and desirable. In this paper, we develop the maximum likelihood learning algorithm and its variation based on short-run Markov chain Monte Carlo sampling from the prior and the posterior distributions of the latent vector, and we show that the learned model exhibits strong performance in terms of image and text generation and anomaly detection. Speakers: Bo Pang, Tian Han, Erik Nijkamp, Song-Chun Zhu, Ying Nian Wu