distributed evidence

Alexander Buchholz (who did his PhD at CREST with Nicolas Chopin), Daniel Ahfock, and my friend Sylvia Richardson published a great paper on the distributed computation of Bayesian evidence in Bayesian Analysis. The setting is one of distributed data from several sources with no communication between them, which relates to consensus Monte Carlo even though model choice has not been particularly studied from that perspective. The authors operate under the assumption of conditionally conjugate models, i.e., the existence of a data augmentation scheme into an exponential family so that conjugate priors can be used. For a division of the data into S blocks, the fundamental identity in the paper is

where α is the normalising constant of the sub-prior exp{log[p(θ)]/S} and the other terms are associated with this prior. Under the conditionally conjugate assumption, the integral can be approximated based on the latent variables. Most interestingly, the associated variance is directly connected with the variance of

under the joint:

“The variance of the ratio measures the quality of the product of the conditional sub-posterior as an importance sample proposal distribution.”

Assuming this variance is finite (which is likely). An approximate alternative is proposed, namely to replace the exact sub-posterior with a Normal distribution, as in consensus Monte Carlo, which should obviously require some consideration as to which parameterisation of the model produces the “most normal” (or the least abnormal!) posterior. And ensures a finite variance in the importance sampling approximation (as ensured by the strong bounds in Proposition 5). A problem shared by the bridgesampling package.

“…if the error that comes from MCMC sampling is relatively small and that the shard sizes are large enough so that the quality of the subposterior normal approximation is reasonable, our suggested approach will result in good approximations of the full data set marginal likelihood.”

The resulting approximation can also be handy in conjunction with reversible jump MCMC, in the sense that RJMCMC algorithms can be run in parallel on different chunks or shards of the entire dataset. Although the computing gain may be reduced by the need for separate approximations.

sandwiching a marginal

When working recently on a paper for estimating the marginal likelihood, I was pointed out this earlier 2015 paper by Roger Grosse, Zoubin Ghahramani and Ryan Adams, which had escaped till now. The beginning of the paper discusses the shortcomings of importance sampling (when simulating from the prior) and harmonic mean (when simulating from the posterior) as solution. And of anNealed importance sampling (when simulating from a sequence, which sequence?!, of targets). The authors are ending up proposing a sequential Monte Carlo or (posterior) particle learning solution. A remark on annealed importance sampling is that there exist both a forward and a backward version for estimating the marginal likelihood, either starting from a simulation from the prior (easy) or from a simulation from the posterior (hard!). As in, e.g., Nicolas Chopin’s thesis, the intermediate steps are constructed from a subsample of the entire sample.

“In this context, unbiasedness can be misleading: because partition function estimates can vary over many orders of magnitude, it’s common for an unbiased estimator to drastically underestimate Ζ with overwhelming probability, yet occasionally return extremely large estimates. (An extreme example is likelihood weighting, which is unbiased, but is extremely unlikely to give an accurate answer for a high-dimensional model.) Unless the estimator is chosen very carefully, the variance is likely to be extremely large, or even infinite.”

One novel aspect of the paper is to advocate for the simultaneous use of different methods and for producing both lower and upper bounds on the marginal p(y) and wait for them to get close enough. It is however delicate to find upper bounds, except when using the dreaded harmonic mean estimator.  (A nice trick associated with reverse annealed importance sampling is that the reverse chain can be simulated exactly from the posterior if associated with simulated data, except I am rather lost at the connection between the actual and simulated data.) In a sequential harmonic mean version, the authors also look at the dangers of using an harmonic mean but argue the potential infinite variance of the weights does not matter so much for log p(y), without displaying any variance calculation… The paper also contains a substantial experimental section that compares the different solutions evoked so far, plus others like nested sampling. Which did not work poorly in the experiment (see below) but could not be trusted to provide a lower or an upper bound. The computing time to achieve some level of agreement is however rather daunting. An interesting read definitely (and I wonder what happened to the paper in the end).

marginal likelihood with large amounts of missing data

In 2018, Panayiota Touloupou, research fellow at Warwick, and her co-authors published a paper in Bayesian analysis that somehow escaped my radar, despite standing in my first circle of topics of interest! They construct an importance sampling approach to the approximation of the marginal likelihood, the importance function being approximated from a preliminary MCMC run, and consider the special case when the sampling density (i.e., the likelihood) can be represented as the marginal of a joint density. While this demarginalisation perspective is rather usual, the central point they make is that it is more efficient to estimate the sampling density based on the auxiliary or latent variables than to consider the joint posterior distribution of parameter and latent in the importance sampler. This induces a considerable reduction in dimension and hence explains (in part) why the approach should prove more efficient. Even though the approximation itself is costly, at about 5 seconds per marginal likelihood. But a nice feature of the paper is to include the above graph that includes both computing time and variability for different methods (the blue range corresponding to the marginal importance solution, the red range to RJMCMC and the green range to Chib’s estimate). Note that bridge sampling does not appear on the picture but returns a variability that is similar to the proposed methodology.

marginal likelihood as exhaustive X validation

In the June issue of Biometrika (for which I am deputy editor) Edwin Fong and Chris Holmes have a short paper (that I did not process!) on the validation of the marginal likelihood as the unique coherent updating rule. Marginal in the general sense of Bissiri et al. (2016). Coherent in the sense of being invariant to the order of input of exchangeable data, if in a somewhat self-defining version (Definition 1). As a consequence, marginal likelihood arises as the unique prequential scoring rule under coherent belief updating in the Bayesian framework. (It is unique given the prior or its generalisation, obviously.)

“…we see that 10% of terms contributing to the marginal likelihood come from out-of-sample predictions, using on average less than 5% of the available training data.”

The paper also contains the interesting remark that the log marginal likelihood is the average leave-p-out X-validation score, across all values of p. Which shows that, provided the marginal can be approximated, the X validation assessment is feasible. Which leads to a highly relevant (imho) spotlight on how this expresses the (deadly) impact of the prior selection on the numerical value of the marginal likelihood. Leaving outsome of the least informative terms in the X-validation leads to exactly the log geometric intrinsic Bayes factor of Berger & Pericchi (1996). Most interesting connection with the Bayes factor community but one that depends on the choice of the dismissed fraction of p‘s.

estimating the marginal likelihood (or an information criterion)

Tory Imai (from Kyoto University) arXived a paper last summer on what first looked like a novel approximation of the marginal likelihood. Based on the variance of thermodynamic integration. The starting argument is that there exists a power 0<t⁰<1 such that the expectation of the logarithm of the product of the prior by the likelihood to the power t⁰ or t⁰-powered likelihood  is equal to the standard log-marginal

when the expectation is under the posterior corresponding to the t⁰-powered likelihood (rather than the full likelihood). By an application of the mean value theorem. Watanabe’s (2013) WBIC replaces the optimum t⁰ with 1/log(n), n being the sample size. The issue in terms of computational statistics is of course that the error of WBIC (against the true log m(x)) is only characterised as an order of n.

The second part of the paper is rather obscure to me, as the motivation for the real log canonical threshold is missing, even though the quantity is connected with the power likelihood. And the DIC effective dimension. It then goes on to propose a new approximation of sBIC, where s stands for singular, of Drton and Plummer (2017) which I had missed (and may ask my colleague Martin later today at Warwick!). Quickly reading through the later however brings explanations about the real log canonical threshold being simply the effective dimension in Schwarwz’s BIC approximation to the log marginal,

(as derived by Watanabe), where m is called the multiplicity of the real log canonical threshold. Both λ and m being unknown, Drton and Plummer (2017) estimate the above approximation in a Bayesian fashion, which leads to a double indexed marginal approximation for a collection of models. Since this thread leads me further and further from a numerical resolution of the marginal estimation, but brings in a different perspective on mixture Bayesian estimation, I will return to this highly  in a later post. The paper of Imai discusses a different numerical approximation to sBIC, With a potential improvement in computing sBIC. (The paper was proposed as a poster to BayesComp 2020, so I am looking forward discussing it with the author.)