More Quantum Speedups for Multiproposal MCMC

Multiproposal Markov chain Monte Carlo (MCMC) algorithms choose from multiple proposals at each iteration in order to sample from challenging target distributions more efficiently. Recent work demonstrates the possibility of quadratic quantum speedups for one such multiproposal MCMC algorithm. Using $P$ proposals, this quantum parallel MCMC (\QP) algorithm requires only $\mathcal{O}(\sqrt{P})$ target evaluations at each step. Here, we present a fast new quantum multiproposal MCMC strategy, \QPP, that only requires $\mathcal{O}(1)$ target evaluations and $\mathcal{O}(\log P)$ qubits. Unlike its slower predecessor, the \QPP\ Markov kernel (\textcolor{red}{1}) maintains detailed balance exactly and (\textcolor{red}{2}) is fully explicit for a large class of graphical models. We demonstrate this flexibility by applying \QPP\ to novel Ising-type models built on bacterial evolutionary networks and obtain significant speedups for Bayesian ancestral trait reconstruction for 248 observed salmonella bacteria.