2016/11/30 by Qiang Liu, Liu, Qiang, Yihao Feng +1 · 3 citations
Computer Science · Physics and Astronomy · Mathematics · #Gaussian Processes and Bayesian Inference #Model Reduction and Neural Networks #Markov Chains and Monte Carlo Methods
paper · pdf · doi:10.48550/arxiv.1612.00081
Variational inference provides a powerful tool for approximate probabilistic in- ference on complex, structured models. Typical variational inference methods, however, require to use inference networks with computationally tractable proba- bility density functions. This largely limits the design and implementation of vari- ational inference methods. We consider wild variational inference methods that do not require tractable density functions on the inference networks, and hence can be applied in more challenging cases. As an example of application, we treat stochastic gradient Langevin dynamics (SGLD) as an inference network, and use our methods to automatically adjust the step sizes of SGLD, yielding significant improvement over the hand-designed step size schemes