Timezone: »

 
Poster
Glow: Generative Flow with Invertible 1x1 Convolutions
Diederik Kingma · Prafulla Dhariwal

Tue Dec 04 07:45 AM -- 09:45 AM (PST) @ Room 210 #11
in Tue Poster Session A »

Flow-based generative models are conceptually attractive due to tractability of the exact log-likelihood, tractability of exact latent-variable inference, and parallelizability of both training and synthesis. In this paper we propose Glow, a simple type of generative flow using invertible 1x1 convolution. Using our method we demonstrate a significant improvement in log-likelihood and qualitative sample quality. Perhaps most strikingly, we demonstrate that a generative model optimized towards the plain log-likelihood objective is capable of efficient synthesis of large and subjectively realistic-looking images.

Keywords: Unsupervised Learning · Nonlinear Dimensionality Reduction and Manifold Learning · Density Estimation · Generative Models · Representation Learning · Latent Variable Models
[ Paper] 

Author Information

Diederik Kingma (Google)
Prafulla Dhariwal (OpenAI)

More from the Same Authors