## likelihood free nested sampling

Posted in Books, Statistics with tags , , , , , , , , , , , on April 26, 2019 by xi'an

A recent paper by Mikelson and Khammash found on bioRxiv considers the (paradoxical?) mixture of nested sampling and intractable likelihood. They however cover only the case when a particle filter or another unbiased estimator of the likelihood function can be found. Unless I am missing something in the paper, this seems a very costly and convoluted approach when pseudo-marginal MCMC is available. Or the rather substantial literature on computational approaches to state-space models. Furthermore simulating under the lower likelihood constraint gets even more intricate than for standard nested sampling as the parameter space is augmented with the likelihood estimator as an extra variable. And this makes a constrained simulation the harder, to the point that the paper need resort to a Dirichlet process Gaussian mixture approximation of the constrained density. It thus sounds quite an intricate approach to the problem. (For one of the realistic examples, the authors mention a 12 hour computation on a 48 core cluster. Producing an approximation of the evidence that is not unarguably stabilised, contrary to the above.) Once again, not being completely up-to-date in sequential Monte Carlo, I may miss a difficulty in analysing such models with other methods, but the proposal seems to be highly demanding with respect to the target.

## dynamic nested sampling for stars

Posted in Books, pictures, Statistics, Travel with tags , , , , , , , , , , , , , , , , , on April 12, 2019 by xi'an

In the sequel of earlier nested sampling packages, like MultiNest, Joshua Speagle has written a new package called dynesty that manages dynamic nested sampling, primarily intended for astronomical applications. Which is the field where nested sampling is the most popular. One of the first remarks in the paper is that nested sampling can be more easily implemented by using a Uniform reparameterisation of the prior, that is, a reparameterisation that turns the prior into a Uniform over the unit hypercube. Which means in fine that the prior distribution can be generated from a fixed vector of uniforms and known transforms. Maybe not such an issue given that this is the prior after all.  The author considers this makes sampling under the likelihood constraint a much simpler problem but it all depends in the end on the concentration of the likelihood within the unit hypercube. And on the ability to reach the higher likelihood slices. I did not see any special trick when looking at the documentation, but reflected on the fundamental connection between nested sampling and this ability. As in the original proposal by John Skilling (2006), the slice volumes are “estimated” by simulated Beta order statistics, with no connection with the actual sequence of simulation or the problem at hand. We did point out our incomprehension for such a scheme in our Biometrika paper with Nicolas Chopin. As in earlier versions, the algorithm attempts at visualising the slices by different bounding techniques, before proceeding to explore the bounded regions by several exploration algorithms, including HMC.

“As with any sampling method, we strongly advocate that Nested Sampling should not be viewed as being strictly“better” or “worse” than MCMC, but rather as a tool that can be more or less useful in certain problems. There is no “One True Method to Rule Them All”, even though it can be tempting to look for one.”

When introducing the dynamic version, the author lists three drawbacks for the static (original) version. One is the reliance on this transform of a Uniform vector over an hypercube. Another one is that the overall runtime is highly sensitive to the choice the prior. (If simulating from the prior rather than an importance function, as suggested in our paper.) A third one is the issue that nested sampling is impervious to the final goal, evidence approximation versus posterior simulation, i.e., uses a constant rate of prior integration. The dynamic version simply modifies the number of point simulated in each slice. According to the (relative) increase in evidence provided by the current slice, estimated through iterations. This makes nested sampling a sort of inversted Wang-Landau since it sharpens the difference between slices. (The dynamic aspects for estimating the volumes of the slices and the stopping rule may hinder convergence in unclear ways, which is not discussed by the paper.) Among the many examples produced in the paper, a 200 dimension Normal target, which is an interesting object for posterior simulation in that most of the posterior mass rests on a ring away from the maximum of the likelihood. But does not seem to merit a mention in the discussion. Another example of heterogeneous regression favourably compares dynesty with MCMC in terms of ESS (but fails to include an HMC version).

[Breaking News: Although I wrote this post before the exciting first image of the black hole in M87 was made public and hence before I was aware of it, the associated AJL paper points out relying on dynesty for comparing several physical models of the phenomenon by nested sampling.]

## 19 dubious ways to compute the marginal likelihood

Posted in Books, Statistics with tags , , , , , , , , , , on December 11, 2018 by xi'an

A recent arXival on nineteen different [and not necessarily dubious!] ways to approximate the marginal likelihood of a given topology of a philogeny tree that reminded me of our San Antonio survey with Jean-Michel Marin. This includes a version of the Laplace approximation called Laplus (!), accounting for the fact that branch lengths on the tree are positive but may have a MAP at zero. Using a Beta, Gamma, or log-Normal distribution instead of a Normal. For importance sampling, the proposals are derived from either the Laplus (!) approximate distributions or from the variational Bayes solution (based on an Normal product). Harmonic means are still used here despite the obvious danger, along with a defensive version that mixes prior and posterior. Naïve Monte Carlo means simulating from the prior, while bridge sampling seems to use samples from prior and posterior distributions. Path and modified path sampling versions are those proposed in 2008 by Nial Friel and Tony Pettitt (QUT). Stepping stone sampling appears like another version of path sampling, also based on a telescopic product of ratios of normalising constants, the generalised version relying on a normalising reference distribution that need be calibrated. CPO and PPD in the above table are two versions based on posterior predictive density estimates.

When running the comparison between so many contenders, the ground truth is selected as the values returned by MrBayes in a massive MCMC experiment amounting to 7.5 billions generations. For five different datasets. The above picture describes mean square errors for the probabilities of split, over ten replicates [when meaningful], the worst case being naïve Monte Carlo, with nested sampling and harmonic mean solutions close by. Similar assessments proceed from a comparison of Kullback-Leibler divergences. With the (predicatble?) note that “the methods do a better job approximating the marginal likelihood of more probable trees than less probable trees”. And massive variability for the poorest methods:

The comparison above does not account for time and since some methods are deterministic (and fast) there is little to do about this. The stepping steps solutions are very costly, while on the middle range bridge sampling outdoes path sampling. The assessment of nested sampling found in the conclusion is that it “would appear to be an unwise choice for estimating the marginal likelihoods of topologies, as it produces poor approximate posteriors” (p.12). Concluding at the Gamma Laplus approximation being the winner across all categories! (There is no ABC solution studied in this paper as the model likelihood can be computed in this setup, contrary to our own setting.)

## computational statistics and molecular simulation [18w5023]

Posted in pictures, Statistics, Travel, University life with tags , , , , , , , , , , , , , , on November 19, 2018 by xi'an

The last day of the X fertilisation workshop at the casa matematicà Oaxaca, there were only three talks and only half of the participants. I lost the subtleties of the first talk by Andrea Agazzi on large deviations for chemical reactions, due to an emergency at work (Warwick). The second talk by Igor Barahona was somewhat disconnected from the rest of the conference, working on document textual analysis by way of algebraic data analysis (analyse des données) methods à la Benzécri. (Who was my office neighbour at Jussieu in the early 1990s.) In the last and final talk, Eric Vanden-Eijden made a link between importance sampling and PDMP, as an integral can be expressed via a trajectory of a path. A generalisation of path sampling, for almost any ODE. But also a competitor to nested sampling, waiting for the path to reach an Hamiltonian level, without some of the difficulties plaguing nested sampling like resampling. And involving continuous time processes. (Is there a continuous time version of ABC as well?!) Returning unbiased estimators of mean (the original integral) and variance. Example of a mixture example in dimension d=10 with k=50 components using only 100 paths.

## subset sampling

Posted in Statistics with tags , , , , , , , , on July 13, 2018 by xi'an

A paper by Au and Beck (2001) was mentioned during a talk at MCqMC 2018 in Rennes and I checked Probabilistic Engineering Mechanics for details. There is no clear indication that the subset simulation advocated therein is particularly effective. The core idea is to obtain the probability to belong to a small set A by a cascading formula, namely the product of the probability to belong to A¹, then the conditional probability to belong to A² given A¹, &tc. When the subsets A¹, A², …, A constitute a decreasing embedded sequence. The simulation conditional on being in one of the subsets $A^i$ is operated by a random-walk Metropolis-within-Gibbs scheme, with an additional rejection when the value is not in the said subset. (Surprisingly, the authors re-establish the validity of this scheme.) Hence the proposal faces similar issues as nested sampling, except that the nested subsets here are defined quite differently as they are essentially free, provided they can be easily evaluated. Each of the random walks need be scaled, the harder a task because this depends on the corresponding subset volume. The subsets $A^i$ themselves are rarely defined in a natural manner, except when being tail events. And need to be calibrated so that the conditional probability of falling into each remains large enough, the cost of free choice. The Markov chain on the previous subset $A^i$ can prove useful to build the next subset $A^{i+1}$, but there is no general principle behind this remark. (If any, this is connected with X entropy.) But else, the past chains are very much wasted, compared with, say, an SMC treatment of the problem. The paper also notices that starting a Markov chain in the set $A^{i+1}$ means there is no burnin time and hence that the probability estimators are thus unbiased. (This creates a correlation between successive Markov chains, but I think it could be ignored if the starting point was chosen at random or after a random number of extra steps.) The authors further point out that the chain may fail to be ergodic, if the proposal distribution lacks energy to link connected regions of the current subset $A^i$. They suggest using multiple chains with multiple starting points, which alleviates the issue only to some extent, as it ultimately depends on the spread of the starting points. As acknowledged in the paper.

## parallelizable sampling method for parameter inference of large biochemical reaction models

Posted in Books, Statistics with tags , , , , , , , , on June 18, 2018 by xi'an

I came across this older (2016) arXiv paper by Jan Mikelson and Mustafa Khammash [antidated as of April 25, 2018] as another version of nested sampling. The novelty of the approach is in applying nested sampling for approximating the likelihood function in the case of involved hidden Markov models (although the name itself does not appear in the paper). This is an interesting proposal, even though there is a fairly large and very active literature on computational approaches to such objects, from sequential Monte Carlo (SMC) to particle MCMC (pMCMC), to SMC².

“We found a way to efficiently sample parameter vectors (particles) from the super level set of the likelihood (sets of particles with a likelihood equal to or higher than some threshold) corresponding to an increasing sequence of thresholds” (p.2)

The approach here is an aggregate of nested sampling and particle filters (SMC), filters that are paradoxically employed in approximating the likelihood function itself, thus called repeatedly as the value of the parameter θ changes, unless I am confused, when it seems to me that, once started with particle filters, the authors could have used them all the way to the upper level (through, again, SMC²). Instead, and that brings a further degree of (uncorrected) approximation to the procedure, a Dirichlet process prior is used to estimate Gaussian mixture approximations to the true posterior distribution(s) on the (super) level sets. Now, approximating a distribution that is zero outside a compact set [the prior restricted to the likelihood being larger than by a distribution with an infinite support does not a priori sound like a particularly enticing idea. Note also that there is no later correction for using the mixture approximation to the restricted prior. (The method also involves an approximation of the (Lebesgue) volume of the level sets that may be poor in higher dimensions.)

“DP-GMM estimations work very well in high dimensional spaces and since we use rejection sampling to obtain samples from the level set by sampling from the DP-GMM estimation, the estimation error does not get propagated through iterations.” (p.13)

One aspect of the paper that puzzles me is the use of a rejection sampler to produce new parameters simulations from a given (super) level set, as this involves a lower bound M on the Gaussian mixture approximation over this level set. If a Gaussian mixture approximation is available, there is apparently no need for this as it can be sampled directly and values below the threshold can be disposed of. It is also unclear why the error does not propagate from one level to the next, if only because of the connection between the successive particle approximations.

## the [not so infamous] arithmetic mean estimator

Posted in Books, Statistics with tags , , , , , , , , , on June 15, 2018 by xi'an

“Unfortunately, no perfect solution exists.” Anna Pajor

Another paper about harmonic and not-so-harmonic mean estimators that I (also) missed came out last year in Bayesian Analysis. The author is Anna Pajor, whose earlier note with Osiewalski I also spotted on the same day. The idea behind the approach [which belongs to the branch of Monte Carlo methods requiring additional simulations after an MCMC run] is to start as the corrected harmonic mean estimator on a restricted set A as to avoid tails of the distributions and the connected infinite variance issues that plague the harmonic mean estimator (an old ‘Og tune!). The marginal density p(y) then satisfies an identity involving the prior expectation of the likelihood function restricted to A divided by the posterior coverage of A. Which makes the resulting estimator unbiased only when this posterior coverage of A is known, which does not seem realist or efficient, except if A is an HPD region, as suggested in our earlier “safe” harmonic mean paper. And efficient only when A is well-chosen in terms of the likelihood function. In practice, the author notes that P(A|y) is to be estimated from the MCMC sequence and that the set A should be chosen to return large values of the likelihood, p(y|θ), through importance sampling, hence missing somehow the double opportunity of using an HPD region. Hence using the same default choice as in Lenk (2009), an HPD region which lower bound is derived as the minimum likelihood in the MCMC sample, “range of the posterior sampler output”. Meaning P(A|y)=1. (As an aside, the paper does not produce optimality properties or even heuristics towards efficiently choosing the various parameters to be calibrated in the algorithm, like the set A itself. As another aside, the paper concludes with a simulation study on an AR(p) model where the marginal may be obtained in closed form if stationarity is not imposed, which I first balked at, before realising that even in this setting both the posterior and the marginal do exist for a finite sample size, and hence the later can be estimated consistently by Monte Carlo methods.) A last remark is that computing costs are not discussed in the comparison of methods.

The final experiment in the paper is aiming at the marginal of a mixture model posterior, operating on the galaxy benchmark used by Roeder (1990) and about every other paper on mixtures since then (incl. ours). The prior is pseudo-conjugate, as in Chib (1995). And label-switching is handled by a random permutation of indices at each iteration. Which may not be enough to fight the attraction of the current mode on a Gibbs sampler and hence does not automatically correct Chib’s solution. As shown in Table 7 by the divergence with Radford Neal’s (1999) computations of the marginals, which happen to be quite close to the approximation proposed by the author. (As an aside, the paper mentions poor performances of Chib’s method when centred at the posterior mean, but this is a setting where the posterior mean is meaningless because of the permutation invariance. As another, I do not understand how the RMSE can be computed in this real data situation.) The comparison is limited to Chib’s method and a few versions of arithmetic and harmonic means. Missing nested sampling (Skilling, 2006; Chopin and X, 2011), and attuned importance sampling as in Berkoff et al. (2003), Marin, Mengersen and X (2005), and the most recent Lee and X (2016) in Bayesian Analysis.