Corpus ID: 233210402

Couplings for Multinomial Hamiltonian Monte Carlo

@inproceedings{Xu2021CouplingsFM,
  title={Couplings for Multinomial Hamiltonian Monte Carlo},
  author={Kai Xu and Tor Erlend Fjelde and Charles Sutton and Hong Ge},
  booktitle={AISTATS},
  year={2021}
}
Hamiltonian Monte Carlo (HMC) is a popular sampling method in Bayesian inference. Recently, Heng & Jacob (2019) studied Metropolis HMC with couplings for unbiased Monte Carlo estimation, establishing a generic parallelizable scheme for HMC. However, in practice a different HMC method, multinomial HMC, is considered as the go-to method, e.g. as part of the no-U-turn sampler. In multinomial HMC, proposed states are not limited to end-points as in Metropolis HMC; instead points along the entire… Expand

Figures and Tables from this paper

References

SHOWING 1-10 OF 48 REFERENCES
Unbiased Hamiltonian Monte Carlo with couplings
Unbiased Markov chain Monte Carlo methods with couplings
2019) again with the momentum variable
  • 2019
A Conceptual Introduction to Hamiltonian Monte Carlo
Coupling of Particle Filters
CODA: convergence diagnosis and output analysis for MCMC
As the leapfrog integrator is of order two (Hairer et al., 2006; Bou-Rabee et al., 2020), for any sufficently small step size ε and number of step l states
  • 2020
Unbiased Contrastive Divergence Algorithm for Training Energy-Based Latent Variable Models
AdvancedHMC.jl: A robust, modular and e cient implementation of advanced HMC algorithms
...
1
2
3
4
5
...