The Zig-Zag Process and Super-Efficient Sampling for Bayesian Analysis of Big Data

@article{Bierkens2016TheZP,
  title={The Zig-Zag Process and Super-Efficient Sampling for Bayesian Analysis of Big Data},
  author={J. Bierkens and P. Fearnhead and G. Roberts},
  journal={arXiv: Computation},
  year={2016}
}
  • J. Bierkens, P. Fearnhead, G. Roberts
  • Published 2016
  • Mathematics
  • arXiv: Computation
  • Standard MCMC methods can scale poorly to big data settings due to the need to evaluate the likelihood at each iteration. There have been a number of approximate MCMC algorithms that use sub-sampling ideas to reduce this computational burden, but with the drawback that these algorithms no longer target the true posterior distribution. We introduce a new family of Monte Carlo methods based upon a multi-dimensional version of the Zig-Zag process of (Bierkens, Roberts, 2017), a continuous time… CONTINUE READING

    Figures from this paper.

    Paper Mentions

    Limit theorems for the Zig-Zag process
    • 24
    • PDF
    The Scalable Langevin Exact Algorithm : Bayesian Inference for Big Data
    • 24
    On Markov chain Monte Carlo methods for tall data
    • 147
    • Highly Influenced
    • PDF
    Piecewise Deterministic Markov Processes for Continuous-Time Monte Carlo
    • 40
    • PDF
    Control variates for stochastic gradient MCMC
    • 47
    • PDF
    Posterior computation with the Gibbs zig-zag sampler
    • 1
    • Highly Influenced
    • PDF
    Informed sub-sampling MCMC: approximate Bayesian inference for large datasets
    • 7
    • PDF
    Scalable Metropolis-Hastings for Exact Bayesian Inference with Large Datasets
    • 6
    • Highly Influenced
    • PDF

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 57 REFERENCES
    The Scalable Langevin Exact Algorithm : Bayesian Inference for Big Data
    • 24
    Monte Carlo Sampling Methods Using Markov Chains and Their Applications
    • 11,923
    • Highly Influential
    • PDF
    A Complete Recipe for Stochastic Gradient MCMC
    • 233
    • PDF
    Irreversible Langevin samplers and variance reduction: a large deviations approach
    • 51
    • PDF
    On Markov chain Monte Carlo methods for tall data
    • 147
    • PDF
    The pseudo-marginal approach for efficient Monte Carlo computations
    • 622
    • PDF