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 Markus Heinonen and Edwin V. Bonilla and Zheyang Shen and Maurizio Filippone},
  journal={ArXiv},
  year={2020},
  volume={abs/2003.03080}
}
  • Simone Rossi, Markus Heinonen, +2 authors Maurizio Filippone
  • Published in ArXiv 2020
  • Mathematics, Computer Science
  • 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… CONTINUE READING

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 51 REFERENCES

    Gaussian Processes for Big Data

    VIEW 9 EXCERPTS
    HIGHLY INFLUENTIAL

    MCMC Using Hamiltonian Dynamics

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    GPflow: A Gaussian Process Library using TensorFlow

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Variational Learning of Inducing Variables in Sparse Gaussian Processes

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL

    Sparse Orthogonal Variational Inference for Gaussian Processes

    VIEW 3 EXCERPTS