delayed-acceptance. ADA boosted

Samuel Wiqvist and co-authors from Scandinavia have recently arXived a paper on a new version of delayed acceptance MCMC. The ADA in the novel algorithm stands for approximate and accelerated, where the approximation in the first stage is to use a Gaussian process to replace the likelihood. In our approach, we used subsets for partial likelihoods, ordering them so that the most varying sub-likelihoods were evaluated first. Furthermore, if a parameter reaches the second stage, the likelihood is not necessarily evaluated, based on the global probability that a second stage is rejected or accepted. Which of course creates an approximation. Even when using a local predictor of the probability. The outcome of a comparison in two complex models is that the delayed approach does not necessarily do better than particle MCMC in terms of effective sample size per second, since it does reject significantly more. Using various types of surrogate likelihoods and assessments of the approximation effect could boost the appeal of the method. Maybe using ABC first could suggest another surrogate?

One Response to “delayed-acceptance. ADA boosted”

  1. Dear professor Robert, I am happy to see that you found our paper interesting! Regarding the comparison between PMCMC and ADA, ADA indeed rejects more proposals than PMCMC, however, we also have that ADA rejects a similar number of proposals as DA (the standard and exact delayed-acceptance scheme). The main comparison of interest is between delayed-acceptance schemes, that is, compare ADA vs DA, not ADA vs PMCMC. We conclude that ADA and DA have similar ESS and acceptance rates, they generate similar inference results while ADA allows for a 2-4 times speed-up in terms of run-time, compared to DA. See Table 3, Table 6 and Figure 5b and 5d.
    Best, Samuel Wiqvist and co-authors

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: