## short course on MCMC at CiRM [slides]

Posted in Statistics with tags , , , , , , , , , , , , , , , on October 23, 2018 by xi'an

Here are the [recycled] slides for the introductory lecture I gave this morning at CIRM, with the side information that it appears Slideshare has gone to another of these stages when slides cannot be played on this blog [when using Firefox]…

## IMS workshop [day 3]

Posted in pictures, R, Statistics, Travel, University life with tags , , , , , , , , , , , , , , , , , , , on August 30, 2018 by xi'an

I made the “capital” mistake of walking across the entire NUS campus this morning, which is quite green and pretty, but which almost enjoys an additional dimension brought by such an intense humidity that one feels having to get around this humidity!, a feature I have managed to completely erase from my memory of my previous visit there. Anyway, nothing of any relevance. oNE talk in the morning was by Markus Eisenbach on tools used by physicists to speed up Monte Carlo methods, like the Wang-Landau flat histogram, towards computing the partition function, or the distribution of the energy levels, definitely addressing issues close to my interest, but somewhat beyond my reach for using a different language and stress, as often in physics. (I mean, as often in physics talks I attend.) An idea that came out clear to me was to bypass a (flat) histogram target and aim directly at a constant slope cdf for the energy levels. (But got scared away by the Fourier transforms!)

Lawrence Murray then discussed some features of the Birch probabilistic programming language he is currently developing, especially a fairly fascinating concept of delayed sampling, which connects with locally-optimal proposals and Rao Blackwellisation. Which I plan to get back to later [and hopefully sooner than later!].

In the afternoon, Maria de Iorio gave a talk about the construction of nonparametric priors that create dependence between a sequence of functions, a notion I had not thought of before, with an array of possibilities when using the stick breaking construction of Dirichlet processes.

And Christophe Andrieu gave a very smooth and helpful entry to partly deterministic Markov processes (PDMP) in preparation for talks he is giving next week for the continuation of the workshop at IMS. Starting with the guided random walk of Gustafson (1998), which extended a bit later into the non-reversible paper of Diaconis, Holmes, and Neal (2000). Although I had a vague idea of the contents of these papers, the role of the velocity ν became much clearer. And premonitory of the advances made by the more recent PDMP proposals. There is obviously a continuation with the equally pedagogical talk Christophe gave at MCqMC in Rennes two months [and half the globe] ago,  but the focus being somewhat different, it really felt like a new talk [my short term memory may also play some role in this feeling!, as I now remember the discussion of Hilderbrand (2002) for non-reversible processes]. An introduction to the topic I would recommend to anyone interested in this new branch of Monte Carlo simulation! To be followed by the most recently arXived hypocoercivity paper by Christophe and co-authors.

## Metropolis-Hastings importance sampling

Posted in Books, Statistics, University life with tags , , , , , , , , , on June 6, 2018 by xi'an

[Warning: As I first got the paper from the authors and sent them my comments, this paper read contains their reply as well.]

In a sort of crazy coincidence, Daniel Rudolf and Björn Sprungk arXived a paper on a Metropolis-Hastings importance sampling estimator that offers similarities with  the one by Ingmar Schuster and Ilja Klebanov posted on arXiv the same day. The major difference in the construction of the importance sampler is that Rudolf and Sprungk use the conditional distribution of the proposal in the denominator of their importance weight, while Schuster and Klebanov go for the marginal (or a Rao-Blackwell representation of the marginal), mostly in an independent Metropolis-Hastings setting (for convergence) and for a discretised Langevin version in the applications. The former use a very functional L² approach to convergence (which reminded me of the early Schervish and Carlin, 1990, paper on the convergence of MCMC algorithms), not all of it necessary in my opinion. As for instance the extension of convergence properties to the augmented chain, namely (current, proposed), is rather straightforward since the proposed chain is a random transform of the current chain. An interesting remark at the end of the proof of the CLT is that the asymptotic variance of the importance sampling estimator is the same as with iid realisations from the target. This is a point we also noticed when constructing population Monte Carlo techniques (more than ten years ago), namely that dependence on the past in sequential Monte Carlo does not impact the validation and the moments of the resulting estimators, simply because “everything cancels” in importance ratios. The mean square error bound on the Monte Carlo error (Theorem 20) is not very surprising as the term ρ(y)²/P(x,y) appears naturally in the variance of importance samplers.

The first illustration where the importance sampler does worse than the initial MCMC estimator for a wide range of acceptance probabilities (Figures 2 and 3, which is which?) and I do not understand the opposite conclusion from the authors.

Indeed the formulation in our paper is unfortunate. The point we want to stress is that we observed in the numerical experiments certain ranges of step-sizes for which MH importance sampling shows a better performance than the classical MH algorithm with optimal scaling. Meaning that the MH importance sampling with optimal step-size can outperform MH sampling, without using additional computational resources. Surprisingly, the optimal step-size for the MH importance sampling estimator seems to remain constant for an increasing dimension in contrast to the well-known optimal scaling of the MH algorithm (given by a constant optimal acceptance rate).

The second uses the Pima Indian diabetes benchmark, amusingly (?) referring to Chopin and Ridgway (2017) who warn against the recourse to this dataset and to this model! The loss in mean square error due to the importance sampling may again be massive (Figure 5) and setting for an optimisation of the scaling factor in Metropolis-Hastings algorithms sounds unrealistic.

Indeed, Chopin and Ridgway suggest more complex problems with a larger number of covariates as benchmarks. However, the well-studied PIMA data set is a sufficient example in order to illustrate the possible benefits but also the limitations of the MH importance sampling approach. The latter are clearly (a) the required knowledge about the optimal step-size—otherwise the performance can indeed be dramatically worse than for the MH algorithm—and (b) the restriction to a small or at most moderate number of covariates. As you are indicating, optimizing the scaling factor is a challenging task. However, the hope is to derive some simple rule of thumb for the MH importance sampler similar to the well-known acceptance rate tuning for the standard MCMC estimator.

## Markov chain importance sampling

Posted in Books, pictures, Running, Statistics, Travel, University life with tags , , , , , , , , , , , on May 31, 2018 by xi'an

Ingmar Schuster (formerly a postdoc at Dauphine and now in Freie Universität Berlin) and Ilja Klebanov (from Berlin) have recently arXived a paper on recycling proposed values in [a rather large class of] Metropolis-Hastings and unadjusted Langevin algorithms. This means using the proposed variates of one of these algorithms as in an importance sampler, with an importance weight going from the target over the (fully conditional) proposal to the target over the marginal stationary target. In the Metropolis-Hastings case, since the later is not available in most setups, the authors suggest using a Rao-Blackwellised nonparametric estimate based on the entire MCMC chain. Or a subset.

“Our estimator refutes the folk theorem that it is hard to estimate [the normalising constant] with mainstream Monte Carlo methods such as Metropolis-Hastings.”

The paper thus brings an interesting focus on the proposed values, rather than on the original Markov chain,  which naturally brings back to mind the derivation of the joint distribution of these proposed values we made in our (1996) Rao-Blackwellisation paper with George Casella. Where we considered a parametric and non-asymptotic version of this distribution, which brings a guaranteed improvement to MCMC (Metropolis-Hastings) estimates of integrals. In subsequent papers with George, we tried to quantify this improvement and to compare different importance samplers based on some importance sampling corrections, but as far as I remember, we only got partial results along this way, and did not cover the special case of the normalising constant Þ… Normalising constants did not seem such a pressing issue at that time, I figure. (A Monte Carlo 101 question: how can we be certain the importance sampler offers a finite variance?)

I think this is interesting future work. My intuition is that for Metropolis-Hastings importance sampling with random walk proposals, the variance is guaranteed to be finite because the importance distribution ρ_θ is a convolution of your target ρ with the random walk kernel q. This guarantees that the tails of ρ_θ are no lighter than those of ρ. What other forms of q mean for the tails of ρ_θ I have less intuition about.

When considering the Langevin alternative with transition (4), I was first confused and thought it was incorrect for moving from one value of Y (proposal) to the next. But that’s what unadjusted means in “unadjusted Langevin”! As pointed out in the early Langevin literature, e.g., by Gareth Roberts and Richard Tweedie, using a discretised Langevin diffusion in an MCMC framework means there is a risk of non-stationarity & non-ergodicity. Obviously, the corrected (MALA) version is more delicate to approximate (?) but at the very least it ensures the Markov chain does not diverge. Even when the unadjusted Langevin has a stationary regime, its joint distribution is likely quite far from the joint distribution of a proper discretisation. Now this also made me think about a parameterised version in the 1996 paper spirit, but there is nothing specific about MALA that would prevent the implementation of the general principle. As for the unadjusted version, the joint distribution is directly available.  (But not necessarily the marginals.)

Personally, I think the most interesting part is the practical performance gain in terms of estimation accuracy for fixed CPU time, combined with the convergence guarantee from the CLT. ULA was particularly important to us because of the papers of Arnak Dalalyan, Alain Durmus & Eric Moulines and recently from Mike Jordan’s group, which all look at an unadjusted Langevin diffusion (and unimodal target distributions). But MALA admits a Metropolis-Hastings importance sampling estimator, just as Random Walk Metropolis does – we didn’t include MALA in the experiments to not get people confused with MALA and ULA. But there is no delicacy involved whatsoever in approximating the marginal MALA proposal distribution. The beauty of our approach is that it works for almost all Metropolis-Hastings algorithms where you can evaluate the proposal density q, there is no constraint to use random walks at all (we will emphasize this more in the paper).

## an interesting identity

Posted in Books, pictures, Statistics, University life with tags , , , , , , , , on March 1, 2018 by xi'an

Another interesting X validated question, another remembrance of past discussions on that issue. Discussions that took place in the Institut d’Astrophysique de Paris, nearby this painting of Laplace, when working on our cosmostats project. Namely the potential appeal of recycling multidimensional simulations by permuting the individual components in nearly independent settings. As shown by the variance decomposition in my answer, when opposing N iid pairs (X,Y) to the N combinations of √N simulations of X and √N simulations of Y, the comparison

$\text{var} \hat{\mathfrak{h}}^2_N=\text{var} (\hat{\mathfrak{h}}^1_N)+\frac{mn(n-1)}{N^2}\,\text{var}^Y\left\{ \mathbb{E}^{X}\left\{\mathfrak{h}(X,Y)\right\}\right\}$

$+\frac{m(m-1)n}{N^2}\,\text{var}^X\left[\mathbb{E}^Y\left\{\mathfrak{h}(X,Y)\right\}\right]$

unsurprisingly gives the upper hand to the iid sequence. A sort of converse to Rao-Blackwellisation…. Unless the production of N simulations gets much more costly when compared with the N function evaluations. No wonder we never see this proposal in Monte Carlo textbooks!

## Russian roulette still rolling

Posted in Statistics with tags , , , , , , , , , , , , on March 22, 2017 by xi'an

Colin Wei and Iain Murray arXived a new version of their paper on doubly-intractable distributions, which is to be presented at AISTATS. It builds upon the Russian roulette estimator of Lyne et al. (2015), which itself exploits the debiasing technique of McLeish et al. (2011) [found earlier in the physics literature as in Carter and Cashwell, 1975, according to the current paper]. Such an unbiased estimator of the inverse of the normalising constant can be used for pseudo-marginal MCMC, except that the estimator is sometimes negative and has to be so as proved by Pierre Jacob and co-authors. As I discussed in my post on the Russian roulette estimator, replacing the negative estimate with its absolute value does not seem right because a negative value indicates that the quantity is close to zero, hence replacing it with zero would sound more appropriate. Wei and Murray start from the property that, while the expectation of the importance weight is equal to the normalising constant, the expectation of the inverse of the importance weight converges to the inverse of the weight for an MCMC chain. This however sounds like an harmonic mean estimate because the property would also stand for any substitute to the importance density, as it only requires the density to integrate to one… As noted in the paper, the variance of the resulting Roulette estimator “will be high” or even infinite. Following Glynn et al. (2014), the authors build a coupled version of that solution, which key feature is to cut the higher order terms in the debiasing estimator. This does not guarantee finite variance or positivity of the estimate, though. In order to decrease the variance (assuming it is finite), backward coupling is introduced, with a Rao-Blackwellisation step using our 1996 Biometrika derivation. Which happens to be of lower cost than the standard Rao-Blackwellisation in that special case, O(N) versus O(N²), N being the stopping rule used in the debiasing estimator. Under the assumption that the inverse importance weight has finite expectation [wrt the importance density], the resulting backward-coupling Russian roulette estimator can be proven to be unbiased, as it enjoys a finite expectation. (As in the generalised harmonic mean case, the constraint imposes thinner tails on the importance function, which then hampers the convergence of the MCMC chain.) No mention is made of achieving finite variance for those estimators, which again is a serious concern due to the similarity with harmonic means…

## recycling Gibbs auxiliaries [a reply]

Posted in Books, pictures, Statistics, University life with tags , , , , , , , on January 3, 2017 by xi'an

[Here is a reply sent to me by Luca Martino, Victor Elvira, and Gustau Camp-Vallis, after my earlier comments on their paper.]

We provide our contribution to the discussion, reporting our experience with the application of Metropolis-within-Gibbs schemes. Since in literature there are miscellaneous opinions, we want to point out the following considerations:

– according to our experience, the use of M>1 steps of the Metropolis-Hastings (MH) method for drawing from each full-conditional (with or without recycling), decreases the MSE of the estimation (see code Ex1-Ex2 and related Figure 7(b) and Figures 8). If the corresponding full conditional is very concentrated, one possible solution is to applied an adaptive or automatic MH for drawing from this full-conditional (it can require the use of M internal steps; see references in Section 3.2).

– Fixing the number of evaluations of the posterior, the comparison between a longer Gibbs chain with a single step of MH and a shorter Gibbs chain with M>1 steps of MH per each full-conditional, is required. Generally, there is no clear winner. The better performance depends on different aspects: the specific scenario, if and adaptive MH is employed or not, if the recycling is applied or not (see Figure 10(a) and the corresponding code Ex2).

The previous considerations are supported/endorsed by several authors (see the references in Section 3.2). In order to highlight the number of controversial opinions about the MH-within-Gibbs implementation, we report a last observation:

– If it is possible to draw directly from the full-conditionals, of course this is the best scenario (this is our belief). Remarkably, as also reported in Chapter 1, page 393 of the book “Monte Carlo Statistical Methods”, C. Robert and Casella, 2004, some authors have found that a “bad” choice of the proposal function in the MH step (i.e., different from the full conditional, or a poor approximation of it) can improve the performance of the MH-within-Gibbs sampler. Namely, they assert that a more “precise” approximation of the full-conditional does not necessarily improve the overall performance. In our opinion, this is possibly due to the fact that the acceptance rate in the MH step (lower than 1) induces an “accidental” random scan of the components of the target pdf in the Gibbs sampler, which can improve the performance in some cases. In our work, for the simplicity, we only focus on the deterministic scan. However, a random scan could be also considered.