Revisiting Reweighted Wake-Sleep for Models with Stochastic Control Flow

Tuan Anh Le, Adam R. Kosiorek, N Siddharth, Yee Whye Teh, Frank Wood

Research output: Chapter in Book/Report/Conference proceedingConference contribution


Stochastic control-flow models (SCFMs) are a class of generative models that involve branching on choices from discrete random variables. Amortized gradient-based learning of SCFMs is challenging as most approaches targeting discrete variables rely on their continuous relaxations—which can be intractable in SCFMs, as branching on relaxations requires evaluating all (exponentially many) branching paths. Tractable alternatives mainly combine REINFORCE with complex control-variate schemes to improve the variance of naive estimators. Here, we revisit the reweighted wake-sleep (RWS) [5] algorithm, and through extensive evaluations, show that it outperforms current state-of-the-art methods in learning SCFMs. Further, in contrast to the importance weighted autoencoder, we observe that RWS learns better models and inference networks with increasing numbers of particles. Our results suggest that RWS is a competitive, often preferable, alternative for learning SCFMs.
Original languageEnglish
Title of host publicationProceedings of the 35th Conference on Uncertainty in Artificial Intelligence
PublisherAssociation for Uncertainty in Artificial Intelligence (AUAI)
Number of pages11
Publication statusPublished - 25 Jul 2019
Event35th Conference on Uncertainty in Artificial Intelligence, UAI 2019 - Tel Aviv, Israel
Duration: 22 Jul 201925 Jul 2019


Conference35th Conference on Uncertainty in Artificial Intelligence, UAI 2019
Abbreviated titleUAI 2019
CityTel Aviv
Internet address

Fingerprint Dive into the research topics of 'Revisiting Reweighted Wake-Sleep for Models with Stochastic Control Flow'. Together they form a unique fingerprint.

Cite this