Adaptive Hamiltonian and Riemann Manifold Monte Carlo

By Ziyun Wang, Shakir Mohamed, Nando de Freitas. 2013

76 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Stochastic Gradient Hamiltonian Monte Carlo (2014, methodology): “Based on the attractive properties of HMC in terms of rapid exploration of the state space, HMC methods have grown in popularity recently (Neal, 2010; Hoffman & Gelman, 2011; Wang et al., 2013).”
Hamiltonian Monte Carlo Without Detailed Balance (2014, methodology): “…& Sutton, 2011), Hilbert space HMC (Beskos et al., 2011), shadow Hamiltonian methods (Izaguirre & Hampton, 2004), parameter adaptation techniques (Wang et al., 2013), Hamiltonian annealed importance sampling (SohlDickstein & Culpepper, 2012), split HMC (Shahbaba et al., 2011), and tempered…”
Bayesian computation: a summary of the current state, and samples backwards and forwards (2015, methodology): “…development of adaptive MALA algorithms in Atchadé (2006) and Marshall and Roberts (2012) and of adaptive HMC and Riemannian Manifold Monte Carlo in Wang et al. (2013) building on the adaptive scaling theory, resulted in a similar drastic mixing improvement as the original Adaptive Metropolis.”
A Markov Jump Process for More Efficient Hamiltonian Monte Carlo (2015, methodology): “tion techniques [16], Hamiltonian annealed importance sampling [17], split HMC [18], tempered trajectories [9], novel discrete time transition rules [10, 19, 20], stochastic gradient variants on HMC [21], HMC for approximate Bayesian computation [22], and new approximate Hamiltonian integrators [23]”
The Geometric Foundations of Hamiltonian Monte Carlo (2014, methodology): “For example, [75] use Bayesian optimization methods to derive an adaptive Hamiltonian Monte Carlo implementation, but they optimize the integrator step size and the number of integrator steps over only a narrow range of values.”
MCMC for Variationally Sparse Gaussian Processes (2015, methodology): “Rather than allow an adaptive (but convergent) scheme as [25], we run the optimization for 30 iterations of 30 samples each, and use the best parameters for a long run of HMC.”