Corpus ID: 2827101

Variational Boosting: Iteratively Refining Posterior Approximations

@article{Miller2017VariationalBI,
  title={Variational Boosting: Iteratively Refining Posterior Approximations},
  author={Andrew C. Miller and N. Foti and Ryan P. Adams},
  journal={ArXiv},
  year={2017},
  volume={abs/1611.06585}
}
We propose a black-box variational inference method to approximate intractable distributions with an increasingly rich approximating class. Our method, termed variational boosting, iteratively refines an existing variational approximation by solving a sequence of optimization problems, allowing the practitioner to trade computation time for accuracy. We show how to expand the variational approximating class by incorporating additional covariance structure and by introducing new components to… Expand
Boosting Variational Inference With Locally Adaptive Step-Sizes
Boosting Variational Inference: an Optimization Perspective
Boosting Variational Inference
Variational Inference via Transformations on Distributions
Unbiased Implicit Variational Inference
Boosting Black Box Variational Inference
2 Variational inference and Gaussian mixtures
Semi-Implicit Variational Inference
Advances in Variational Inference
Practical Posterior Error Bounds from Variational Objectives
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 47 REFERENCES
Boosting Variational Inference
Nonparametric variational inference
Hierarchical Variational Models
Variational Inference with Normalizing Flows
Black Box Variational Inference
Auto-Encoding Variational Bayes
Gaussian Covariance and Scalable Variational Inference
  • M. Seeger
  • Mathematics, Computer Science
  • ICML
  • 2010
Operator Variational Inference
Stochastic variational inference
...
1
2
3
4
5
...