Rényi Divergence Variational Inference
This work provides a unified framework for variational inference, potentially benefiting researchers and practitioners in machine learning, though it appears incremental as it builds on existing divergence concepts.
The paper tackles the problem of variational inference by introducing the variational Rényi bound, which extends traditional methods to Rényi's alpha-divergences, enabling smooth interpolation from evidence lower-bound to log likelihood. Experiments on Bayesian neural networks and variational auto-encoders demonstrate its wide applicability.
This paper introduces the variational Rényi bound (VR) that extends traditional variational inference to Rényi's alpha-divergences. This new family of variational methods unifies a number of existing approaches, and enables a smooth interpolation from the evidence lower-bound to the log (marginal) likelihood that is controlled by the value of alpha that parametrises the divergence. The reparameterization trick, Monte Carlo approximation and stochastic optimisation methods are deployed to obtain a tractable and unified framework for optimisation. We further consider negative alpha values and propose a novel variational inference method as a new special case in the proposed framework. Experiments on Bayesian neural networks and variational auto-encoders demonstrate the wide applicability of the VR bound.