Corpus ID: 212628377

Rethinking Sparse Gaussian Processes: Bayesian Approaches to Inducing-Variable Approximations

@article{Rossi2020RethinkingSG,
  title={Rethinking Sparse Gaussian Processes: Bayesian Approaches to Inducing-Variable Approximations},
  author={Simone Rossi and M. Heinonen and Edwin V. Bonilla and Zheyang Shen and M. Filippone},
  journal={ArXiv},
  year={2020},
  volume={abs/2003.03080}
}
Variational inference techniques based on inducing variables provide an elegant framework for scalable posterior estimation in Gaussian process (GP) models. Most previous works treat the locations of the inducing variables, i.e. the inducing inputs, as variational hyperparameters, and these are then optimized together with GP covariance hyper-parameters. While some approaches point to the benefits of a Bayesian treatment of GP hyper-parameters, this has been largely overlooked for the inducing… Expand

References

SHOWING 1-10 OF 51 REFERENCES
Generic Inference in Latent Gaussian Process Models
  • 20
  • PDF
Sparse Orthogonal Variational Inference for Gaussian Processes
  • 14
  • PDF
Inference in Deep Gaussian Processes using Stochastic Gradient Hamiltonian Monte Carlo
  • 45
  • Highly Influential
  • PDF
MCMC for Variationally Sparse Gaussian Processes
  • 87
  • PDF
Doubly Stochastic Variational Inference for Deep Gaussian Processes
  • 193
  • PDF
Random Feature Expansions for Deep Gaussian Processes
  • 86
  • PDF
Sparse Gaussian Processes using Pseudo-inputs
  • 1,312
  • PDF
Variational Learning of Inducing Variables in Sparse Gaussian Processes
  • M. Titsias
  • Mathematics, Computer Science
  • AISTATS
  • 2009
  • 866
  • Highly Influential
  • PDF
Implicit Posterior Variational Inference for Deep Gaussian Processes
  • 15
  • Highly Influential
  • PDF
...
1
2
3
4
5
...