stratified ABC [One World ABC webinar]

The third episode of the One World ABC seminar (Season 1!) was kindly delivered by Umberto Picchini on Stratified sampling and bootstrapping for ABC which I already if briefly discussed after BayesComp 2020. Which sounds like a million years ago… His introduction on the importance of estimating the likelihood using a kernel, while 600% justified wrt his talk, made the One World ABC seminar sounds almost like groundhog day!  The central argument is in the computational gain brought by simulating a single θ dependent [expensive] dataset followed by [cheaper] bootstrap replicates. Which turns de fact into bootstrapping the summary statistics.

If I understand correctly, the post-stratification approach of Art Owen (2013?, I cannot find the reference) corrects a misrepresentation of mine. Indeed, defining a partition with unknown probability weights seemed to me to annihilate the appeal of stratification, because the Bernoulli variance of the estimated probabilities brought back the same variability as the mother estimator. But with bootstrap, this requires only two simulations, one for the weights and one for the target. And further allows for a larger ABC tolerance in fine. Free lunch?!

The speaker in two weeks (21 May or Ascension Thursday!) is my friend and co-author Gael Martin from Monash University, who will speak on Focused Bayesian prediction, at quite a late time down under..!

Computing Bayes: Bayesian Computation from 1763 to the 21st Century

Last night, Gael Martin, David Frazier (from Monash U) and myself arXived a survey on the history of Bayesian computations. This project started when Gael presented a historical overview of Bayesian computation, then entitled ‘Computing Bayes: Bayesian Computation from 1763 to 2017!’, at ‘Bayes on the Beach’ (Queensland, November, 2017). She then decided to build a survey from the material she had gathered, with her usual dedication and stamina. Asking David and I to join forces and bring additional perspectives on this history. While this is a short and hence necessary incomplete history (of not everything!), it hopefully brings some different threads together in an original enough fashion (as I think there is little overlap with recent surveys I wrote). We welcome comments about aspects we missed, skipped or misrepresented, most obviously!

Nested Sampling SMC [a reply]

Here is a response from Robert Salomone following my comments of the earlier day (and pointing out I already commented the paper two years ago):

You may be interested to know that we are at the tail end of carrying out a major revision of the paper, which we hope will be done in the near future — there will be some new theory (we are in the final stages for a consistency proof of the ANS-SMC algorithm with new co-author Adam Johansen), as well as new numerics (including comparisons to Nested Sampling), and additional discussion that clarifies the overall narrative.

A few comments relating your post that may clear some things up:

• The method you describe with the auxiliary variable is actually one of three proposed algorithms. We call this one “Improved Nested Sampling” as it is the algorithm most similar to the original Nested Sampling. Two further extensions are the adaptive SMC sampler, and the fixed SMC sampler – the latter of which is provably consistent and unbiased for the model evidence (we also often see improvements over standard NS for similar computational effort when MCMC is used).
• Regarding computational effort – it is the same for Improved NS (in fact, you can obtain the standard Nested Sampling evidence estimate from the same computational run!). For the adaptive variant, the computational effort is roughly the same for ρ = e⁻¹. In the current version of the paper this is only discussed briefly (last page of p.23). However, in the revision we will include additional experiments comparing the practical performance.
• Regarding the question of “why not regular SMC”; we chose to focus more on why SMC is a good way to do Nested Sampling rather than why Nested Sampling is a good way to do SMC. Our main priority was to show there is a lot of opportunity to develop new nested sampling style algorithms by approaching it from a different angle. That said, Nested Sampling’s primary advantage over standard SMC seems to be in problems involving “phase transitions’’ such as our first example, for which temperature based methods are inherently ill-suited (and will often fail to detect so!).

research position at Monash

My friends (and coauthors) Gael Martin and David Frazier forwarded me this call for a two-year research research fellow position at Monash. Working with the team there on the most exciting topic of approximate Bayes!

The Research Fellow will conduct research associated with ARC Discovery Grant DP200101414: “Loss-Based Bayesian Prediction”. This project proposes a new paradigm for prediction. Using state-of-the-art computational methods, the project aims to produce accurate, fit for purpose predictions which, by design, reduce the loss incurred when the prediction is inaccurate. Theoretical validation of the new predictive method is an expected outcome, as is extensive application of the method to diverse empirical problems, including those based on high-dimensional and hierarchical data sets. The project will exploit recent advances in Bayesian computation, including approximate Bayesian computation and variational inference, to produce predictive distributions that are expressly designed to yield accurate predictions in a given loss measure. The Research Fellow would be expected to engage in all aspects of the research and would therefore build expertise in the methodological, theoretical and empirical aspects of this new predictive approach.

Deadline is 13 May 2020. This is definitely an offer to consider!

Scott Sisson’s ABC seminar in Paris [All about that Bayes]

On the “All about that Bayes” seminar tomorrow (Tuesday 21 at 3p.m., room 42, AgroParisTech, 16 rue Claude Bernard, Paris 5ième), Scott Sisson, School of Mathematics and Statistics at UNSW, and visiting Paris-Dauphine this month, will give a talk on

Approximate posteriors and data for Bayesian inference

Abstract
For various reasons, including large datasets and complex models, approximate inference is becoming increasingly common. In this talk I will provide three vignettes of recent work. These cover a) approximate Bayesian computation for Gaussian process density estimation, b) likelihood-free Gibbs sampling, and c) MCMC for approximate (rounded) data.

Australia’s burning and it ain’t gonna stop anytime soon…

freedom for women jailed in Iran

Three women are currently on hunger strike in Iranian jails to protest against their arbitrary detention under espionage charges and the dire conditions of these detentions. Two of them, Fariba Adelkhah and Nazanin Zaghari-Ratcliffe, are dual nationals with Iranian nationality, whose second nationality is not recognised by Iranian authorities and makes their release from jail the more unlikely. On the French side, Fariba Adelkhah, anthropologist, along with Roland Marchal, sociologist, is a researcher at Sciences Po Paris. They were arrested in June 2019. On the British side, Nazanin Zaghari-Ratcliffe (jailed in April 2016) is working with the Thomson Reuters Foundation, the Canadian news agency Thomson Reuters‘ charitable arm and Kylie Moore-Gilbert (jailed since October 2018) is a  Middle East Lecturer at the University of Melbourne‘s Asia Institute. Petitions have been launched to support them, but I wonder at the possibility to move the Iranian cynical stance other than through the respective governments of these women. (Obviously, there are many many other women and men jailed in Iran for political or other discriminatory reasons who should equally be freed.)