Handbook of Markov Chain Monte Carlo: Hardcover: 619 pages Publisher: Chapman and Hall/CRC Press (first edition, May 2011) Language: English ISBN-10: 1420079417
By Steve P. Brooks, Andrew Gelman, Galin L. Jones, Xiao-Li Meng. 2012
1,845 citations recorded by the source corpus.
Hamiltonian dynamics can be used to produce distant proposals for the Metropolis algorithm, thereby avoiding the slow exploration of the state space that results from the diffusive behaviour of simple random-walk proposals. Though originating in physics, Hamiltonian dynamics can be applied to most problems with continuous state spaces by simply introducing fictitious “momentum” variables. A key to its usefulness is that Hamiltonian dynamics preserves volume, and its trajectories can thus be used to define complex mappings without the need to account for a hard-to-compute Jacobian factor — a property that can be exactly maintained even when the dynamics is approximated by discretizing time. In this review, I discuss theoretical and practical aspects of Hamiltonian Monte Carlo, and present some of its variations, including using windows of states for deciding on acceptance or rejection, computing trajectories using fast approximations, tempering during the course of a trajectory to handle isolated modes, and short-cut methods that prevent useless trajectories from taking much computation time.
Subfields: Artificial Intelligence, Mathematical Physics, Statistics and Probability, Probabilistic & Bayesian Methods
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Chunyuan Li
- Lawrence Carin
- Zhe Gan
- Jianfeng Gao
- Yulia Tsvetkov
- Gilles Louppe
- Yunchen Pu
- Yoshua Bengio
In Mathematical Physics
- Gilles Louppe
- Samuel Olorunfemi Adams
- Michael Betancourt
- Zhe Gan
- Lawrence Carin
- Changyou Chen
- Anthony N. Pettitt
- Lawrence M. Murray
In Statistics and Probability
- Chunyuan Li
- Eike C. Brechmann
- Lutz F. Gruber
- Gilles Louppe
- Changyou Chen
- Lawrence Carin
- Yunchen Pu
- Claudia Czado
In Probabilistic & Bayesian Methods
- Gilles Louppe
- Kyle Cranmer
- Atilim Gunes Baydin
- Frank D. Wood
- Andreas Munk
- Chunyuan Li
- Bradley Gram-Hansen
- Zhe Gan
Citations in the corpus
- Gaussian mixture models with Wasserstein distance (2018, methodology): “To avoid the problem of back propagating through discrete variables, where the reparameterization trick is not possible, we choose to compute the expectation over the discrete distribution qD D (k|x) exactly, as MNIST modeling only requires a small number of classes making this expectation tractable”
- Spatial and spatio-temporal models with R-INLA. (2013, methodology): “Markov Chain Monte Carlo (MCMC) methods (Brooks et al., 2011; Robert and Casella, 2004), are normally used for Bayesian computation, arguably thanks to the wide popularity of the BUGS software (Lunn et al., 2009, 2012).”
- A Review of Modern Computational Algorithms for Bayesian Optimal Design (2016, methodology): “ABC has commonly been used to perform inference (e.g. Drovandi and Pettitt, 2011; Drovandi et al., 2011; Sisson and Fan, 2011).”
- A Practical Sequential Stopping Rule for High-Dimensional Markov Chain Monte Carlo (2016, background): “We direct interested readers to Brooks et al. (2010) for more information.”
- Learning Weight Uncertainty with Stochastic Gradient MCMC for Shape Classification (2016, methodology): “For detailed concepts in MCMC, we suggest [1].”
- Unbiased Markov chain Monte Carlo methods with couplings (2020, background): “Markov chain Monte Carlo (MCMC) methods constitute a popular class of algorithms to approximate high dimensional integrals arising in statistics and other fields (Liu, 2008; Robert and Casella, 2004; Brooks et al., 2011; Green et al., 2015).”