Timezone: »

 
Poster
Learning Latent Space Energy-Based Prior Model
Bo Pang · Tian Han · Erik Nijkamp · Song-Chun Zhu · Ying Nian Wu

Thu Dec 10 09:00 AM -- 11:00 AM (PST) @ Poster Session 5 #1363
in Poster Session 6 »

We propose an energy-based model (EBM) in the latent space of a generator model, so that the EBM serves as a prior model that stands on the top-down network of the generator model. Both the latent space EBM and the top-down network can be learned jointly by maximum likelihood, which involves short-run MCMC sampling from both the prior and posterior distributions of the latent vector. Due to the low dimensionality of the latent space and the expressiveness of the top-down network, a simple EBM in latent space can capture regularities in the data effectively, and MCMC sampling in latent space is efficient and mixes well. We show that the learned model exhibits strong performances in terms of image and text generation and anomaly detection. The one-page code can be found in supplementary materials.

Keywords: Algorithms · Computational Complexity · Algorithms -> Active Learning; Theory

Author Information

Bo Pang (UCLA)
Tian Han (Stevens Institute of Technology)
Erik Nijkamp (UCLA)
Song-Chun Zhu (UCLA)
Ying Nian Wu (University of California, Los Angeles)

More from the Same Authors