Markov chain Monte Carlo without evaluating the target: an auxiliary variable approach

W Yuan, G Wang - arxiv preprint arxiv:2406.05242, 2024 - arxiv.org
In sampling tasks, it is common for target distributions to be known up to a normalising
constant. However, in many situations, evaluating even the unnormalised distribution can be …

A Separation in Heavy-Tailed Sampling: Gaussian vs. Stable Oracles for Proximal Samplers

Y He, A Mousavi-Hosseini, K Balasubramanian… - arxiv preprint arxiv …, 2024 - arxiv.org
We study the complexity of heavy-tailed sampling and present a separation result in terms of
obtaining high-accuracy versus low-accuracy guarantees ie, samplers that require only $ O …

On the large deviation principle for Metropolis-Hastings Markov Chains: the Lyapunov function condition and examples

F Milinanni, P Nyquist - arxiv preprint arxiv:2403.08691, 2024 - arxiv.org
With an aim to analyse the performance of Markov chain Monte Carlo (MCMC) methods, in
our recent work we derive a large deviation principle (LDP) for the empirical measures of …

Alternative representation of the large deviation rate function and hyperparameter tuning schemes for Metropolis-Hastings Markov Chains

F Milinanni - arxiv preprint arxiv:2409.20337, 2024 - arxiv.org
Markov chain Monte Carlo (MCMC) methods are one of the most common classes of
algorithms to sample from a target probability distribution $\pi $. A rising trend in recent …