Timezone: »

Semi-Autoregressive Energy Flows: Exploring Likelihood-Free Training of Normalizing Flows
Phillip Si · Zeyi Chen · Subham S Sahoo · Yair Schiff · Volodymyr Kuleshov

Tue Jul 25 02:00 PM -- 04:30 PM (PDT) @ Exhibit Hall 1 #213

Training normalizing flow generative models can be challenging due to the need to calculate computationally expensive determinants of Jacobians. This paper studies the likelihood-free training of flows and proposes the energy objective, an alternative sample-based loss based on proper scoring rules. The energy objective is determinant-free and supports flexible model architectures that are not easily compatible with maximum likelihood training, including semi-autoregressive energy flows, a novel model family that interpolates between fully autoregressive and non-autoregressive models. Energy flows feature competitive sample quality, posterior inference, and generation speed relative to likelihood-based flows; this performance is decorrelated from the quality of log-likelihood estimates, which are generally very poor. Our findings question the use of maximum likelihood as an objective or a metric, and contribute to a scientific study of its role in generative modeling. Code is available at https://github.com/ps789/SAEF.

Author Information

Phillip Si (Cornell University)
Zeyi Chen (Tsinghua University, Tsinghua University)
Subham S Sahoo (Indian Institute of Technology)

I'm currently an undergraduate student at Indian Institute of Technology - Kharagpur pursuing my major in Electrical Engineering and a minor in Computer Science.

Yair Schiff (Department of Computer Science, Cornell University)
Yair Schiff

I'm a second year PhD student in the Computer Science department at Cornell University. I have also worked as a software engineer at IBM and collaborated with the Trusted AI department in IBM Research. Prior to joining IBM, I completed a MS in Computer Science at Courant Institute at NYU and a BA in Economics at the University of Pennsylvania.

Volodymyr Kuleshov (Cornell Tech)

More from the Same Authors