Bayesian nonparametric models provide a principled way to automatically adapt the complexity of a model to the amount of the data available, but computation in such models is difficult. Amortized variational approximations are appealing because of their computational efficiency, but current methods rely on a fixed finite truncation of the infinite model. This truncation level can be difficult to set, and also interacts poorly with amortized methods due to the over-pruning problem. Instead, we propose a new variational approximation, based on a method from statistical physics called Russian roulette sampling. This allows the variational distribution to adapt its complexity during inference, without relying on a fixed truncation level, and while still obtaining an unbiased estimate of the gradient of the original variational objective. We demonstrate this method on infinite sized variational auto-encoders using a Beta-Bernoulli (Indian buffet process) prior.
Kai Xu (University of Edinburgh)
Akash Srivastava (MIT-IBM, University Of Edinburgh)
Charles Sutton (Google)
Related Events (a corresponding poster, oral, or spotlight)
2019 Poster: Variational Russian Roulette for Deep Bayesian Nonparametrics »
Wed Jun 12th 06:30 -- 09:00 PM Room Pacific Ballroom