Search Results for author: Matti Vihola

Found 13 papers, 11 papers with code

On the convergence of dynamic implementations of Hamiltonian Monte Carlo and No U-Turn Samplers

no code implementations7 Jul 2023 Alain Durmus, Samuel Gruffaz, Miika Kailas, Eero Saksman, Matti Vihola

Under conditions similar to the ones existing for HMC, we also show that NUTS is geometrically ergodic.

Simulating counterfactuals

1 code implementation27 Jun 2023 Juha Karvanen, Santtu Tikka, Matti Vihola

Counterfactual inference considers a hypothetical intervention in a parallel world that shares some evidence with the factual world.

counterfactual Counterfactual Inference +2

bssm: Bayesian Inference of Non-linear and Non-Gaussian State Space Models in R

1 code implementation21 Jan 2021 Jouni Helske, Matti Vihola

We present an R package bssm for Bayesian non-linear/non-Gaussian state space modelling.

Bayesian Inference Computation

Conditional particle filters with diffuse initial distributions

1 code implementation26 Jun 2020 Santeri Karppinen, Matti Vihola

The method only requires simulatable Markov transitions that are reversible with respect to the initial distribution, which can be improper.

Computation Methodology

On the use of approximate Bayesian computation Markov chain Monte Carlo with inflated tolerance and post-correction

2 code implementations1 Feb 2019 Matti Vihola, Jordan Franks

Approximate Bayesian computation allows for inference of complicated probabilistic models with intractable likelihoods using model simulations.

Computation

Graphical model inference: Sequential Monte Carlo meets deterministic approximations

2 code implementations NeurIPS 2018 Fredrik Lindsten, Jouni Helske, Matti Vihola

Approximate inference in probabilistic graphical models (PGMs) can be grouped into deterministic methods and Monte-Carlo-based methods.

Unbiased inference for discretely observed hidden Markov model diffusions

no code implementations26 Jul 2018 Neil K. Chada, Jordan Franks, Ajay Jasra, Kody J. H. Law, Matti Vihola

The resulting estimator leads to inference without a bias from the time-discretisation as the number of Markov chain iterations increases.

Bayesian Inference Methodology Probability Computation 65C05 (primary), 60H35, 65C35, 65C40 (secondary)

Coupled conditional backward sampling particle filter

1 code implementation15 Jun 2018 Anthony Lee, Sumeetpal S. Singh, Matti Vihola

This complements the earlier findings in the literature for conditional particle filters, which assume the number of particles to grow (super)linearly in terms of the time horizon.

Computation Probability Primary 65C05, secondary 60J05, 65C35, 65C40

Importance sampling type estimators based on approximate marginal MCMC

1 code implementation8 Sep 2016 Matti Vihola, Jouni Helske, Jordan Franks

We consider importance sampling (IS) type weighted estimators based on Markov chain Monte Carlo (MCMC) targeting an approximate marginal of the target distribution.

Computation Probability

Unbiased estimators and multilevel Monte Carlo

1 code implementation3 Dec 2015 Matti Vihola

Under general conditions, essentially when MLMC admits the canonical square root Monte Carlo error rate, the proposed new schemes are shown to be asymptotically as efficient as MLMC, both in terms of variance and cost.

Computation Probability 65C05 (Primary), 65C30 (Secondary)

Adaptive parallel tempering algorithm

2 code implementations4 May 2012 Blazej Miasojedow, Eric Moulines, Matti Vihola

Parallel tempering is a generic Markov chain Monte Carlo sampling method which allows good mixing with multimodal target distributions, where conventional Metropolis-Hastings algorithms often fail.

Computation

Robust adaptive Metropolis algorithm with coerced acceptance rate

2 code implementations19 Nov 2010 Matti Vihola

This paper introduces a new robust adaptive Metropolis algorithm estimating the shape of the target distribution and simultaneously coercing the acceptance rate.

Computation 65C40 (Primary) 60J22, 60J05, 93E35 (Secondary)

Grapham: Graphical Models with Adaptive Random Walk Metropolis Algorithms

1 code implementation25 Nov 2008 Matti Vihola

Recently developed adaptive Markov chain Monte Carlo (MCMC) methods have been applied successfully to many problems in Bayesian statistics.

Computation

Cannot find the paper you are looking for? You can Submit a new open access paper.