Vectorized univariate slice sampler

Usage

slice_sample(
  log_prob_fn,
  init,
  n_draws,
  warmup = 200L,
  thin = 10L,
  width = 1,
  max_steps = 100L,
  verbose = FALSE
)

log_prob_fn: Vectorized log-density: takes an m x dim matrix, returns m log-densities (-Inf allowed).
init: n_chains x dim matrix of starting points, all with finite log-density.
n_draws: Number of retained draws in total, across all chains.
warmup: Steps discarded at the start of each chain.
thin: Keep one draw in thin.
width: Initial slice width per dimension (recycled). Adapted during warmup and then held fixed.
max_steps: Cap on stepping-out expansions per coordinate.
verbose: Report progress.

A list with draws (n_draws x dim), chains (n_kept x n_chains x dim array) and n_evals.