Adaptive proposal distribution for random walk Metropolis algorithm

By Heikki Haario, Eero Saksman, Johanna Tamminen. 1999

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

Gradient-free Hamiltonian Monte Carlo with Efficient Kernel Exponential Families (2015, methodology): “For example, in Pseudo-Marginal MCMC (PM-MCMC) [2, 3], the posterior does not have an analytic expression, but can only be estimated at any given point, e.g. in Bayesian Gaussian Process classiﬁcation [4].”
Kernel Adaptive Metropolis-Hastings (2013, methodology): “Adaptiv e MCMC samplers were first studied by Haario et al.(1999; Proceedings of the31 st International Conference on Machine Learning, Beijing, China, 2014. JMLR: W&CP volume 32. Copyright 2014 by the author(s). 2001), where the authors propose to update the proposal distribution along the sampling”
Adaptive Multiple Importance Sampling (2009, methodology): “sive and an involved recycling of the whole set of simulations is then relevant. 6 A banana shape target example This evaluation of the performances of AMIS resorts to the benchmark target density of Haario et al. (1999, 2001), which can be calibrated as to become extremely challenging. The target d”
High‐dimensional posterior exploration of hydrologic models using multiple‐try DREAM(ZS) and high‐performance computing (2012, methodology): “An example of this includes the single-chain AP [ Haario et al. , 1999], AM [ Haario et al. , 2001] and DRAM methods [ Haario et al. , 2006].”
Accelerating Markov Chain Monte Carlo Simulation by Differential Evolution with Self-Adaptive Randomized Subspace Sampling (2009, methodology): “The second case study considers a 10-dimensional twiste d Gaussian density function first introduced in [15] which is given by the unnormalized d ensity πb(x) ∝ π(φb(x)), with ) ,.”
Markov chain Monte Carlo simulation using the DREAM software package: Theory, concepts, and MATLAB implementation (2016, methodology): “The most common adaptive single chain methods are the adaptive proposal (AP) (Haario et al., 1999), adaptive Metropolis (AM) (Haario et al.”
Boosting Variational Inference (2016, methodology): “We demonstrate this property on the “banana” distribution, which is widely-used for studying the performance of MCMC algorithms [Haario et al., 1999, 2001, Roberts and Rosenthal, 2009].”
Efficient MCMC for Climate Model Parameter Estimation: Parallel Adaptive Chains and Early Rejection (2012, background): “Often, such proposals perform well also for mildly non-Gaussian targets, see e.g. the synthetic examples of Haario et al. (1999) and the application studies of Malve et al. (2005); Vahteristo et al. (2009).”
Confronting uncertainty in model-based geostatistics using Markov Chain Monte Carlo simulation (2011, methodology): “Examples of such methods are the Adaptive Proposal (AP) and Adaptive Metropolis (AM) schemes presented in (Haario et al., 1999; Haario et al., 2001), respectively.”
DRAM: Efficient adaptive MCMC (2006, methodology): “In this paper we propose various strategies to combine two quite powerful ideas that have recently appeared in the MCMC literature: adaptive Metropolis samplers [8] [9] and delayed rejection [18] [6] [13].”
Convergence of adaptive mixtures of importance sampling schemes (2007, unknown): “For instance, in the MCMC community, sequential changes in the variance of Markov kernels have been proposed in [13, 14], while adaptive changes taking advantage of regeneration properties of the kernels have been constructed by Gilks, Roberts and Sahu [12] and Sahu and Zhigljavsky [23, 24].”
Adaptively Scaling the Metropolis Algorithm Using Expected Squared Jumped Distance (2007, methodology): “In high-dimensional problems, we have optimized the scale of the proposal kernel while updating the covariance matrix using empirical weighting of posterior simulations (as in Haario et al. (1999)). We also anticipate that these methods can be generalized to optimize over more general MCMC algorithm”