Hi all, Has an adaptive step-size numerical method been considered in place of Leapfrog? There is little in the literature on this but there is one paper ( https://www.jstage.jst.go.jp/article/jsiaml/9/0/9_33/_article ) on the topic based on Hairer’s 2005 paper (Explicit, Time Reversible, Adaptive …

Adaptive step-size method in place of Leapfrog

LJDevlin June 14, 2019, 10:05am 3

Interesting, thanks for the insight Michael.

Topic		Replies	Views
Details on how Stan adaptively tunes the HMC parameters? (i.e. mass matrix, step size and leapfrog steps) General algorithms	4	684	February 20, 2023
Conceptual Questions about HMC and Leapfrog Integrator General	5	719	December 21, 2022
Adaptation of timestep in pure HMC Algorithms	7	854	February 28, 2020
HMC (jittered) vs. NUTS on 1000-dimensional standard normal Algorithms mcmc	9	3882	April 29, 2019
NUTS misses U-turns, runs in circles until max_treedepth Algorithms	66	5512	August 31, 2019