In a simple question on X validated a few days ago [about simulating from x²φ(x)] popped up the remark that the person asking the question wanted a direct simulation method for higher efficiency. Compared with an accept-reject solution. Which shows a misunderstanding of what “efficiency” means on Monte Carlo situations. If it means anything, I would think it is reflected in the average time taken to return one simulation and possibly in the worst case. But there is no reason to call an inverse cdf method more efficient than an accept reject or a transform approach since it all depends on the time it takes to make the inversion compared with the other solutions… Since inverting the closed-form cdf in this example is much more expensive than generating a Gamma(½,½), and taking plus or minus its root, this is certainly the case here. Maybe a ziggurat method could be devised, especially since x²φ(x)<φ(x) when |x|≤1, but I am not sure it is worth the effort!
Archive for Monte Carlo Statistical Methods
Steffen Dereich, WWU Münster, Germany
Paul Dupuis, Brown University, Providence, USA
Mark Girolami, Imperial College London, UK
Emmanuel Gobet, École Polytechnique, Palaiseau, France
Aicke Hinrichs, Johannes Kepler University, Linz, Austria
Alexander Keller, NVIDIA Research, Germany
Gunther Leobacher, Johannes Kepler University, Linz, Austria
Art B. Owen, Stanford University, USA
Note that, while special sessions are already selected, including oneon Stochastic Gradient methods for Monte Carlo and Variational Inference, organised by Victor Elvira and Ingmar Schuster (my only contribution to this session being the suggestion they organise it!), proposals for contributed talks will be selected based on one-page abstracts, to be submitted by March 1.
In answering a simple question on X validated about producing Monte Carlo estimates of the variance of estimators of exp(-θ) in a Poisson model, I wanted to illustrate the accuracy of these estimates against the theoretical values. While one case was easy, since the estimator was a Binomial B(n,exp(-θ)) variate [in yellow on the graph], the other one being the exponential of the negative of the Poisson sample average did not enjoy a closed-form variance and I instead used a first order (δ-method) approximation for this variance which ended up working surprisingly well [in brown] given that the experiment is based on an n=20 sample size.
Thanks to the comments of George Henry, I stand corrected: the variance of the exponential version is easily manageable with two lines of summation! As
which allows for a comparison with its second order Taylor approximation:
We have been working towards a revision of our reparameterisation paper for quite a while now and too advantage of Kate Lee visiting Paris this fortnight to make a final round: we have now arXived (and submitted) the new version. The major change against the earlier version is the extension of the approach to a large class of models that include infinitely divisible distributions, compound Gaussian, Poisson, and exponential distributions, and completely monotonic densities. The concept remains identical: change the parameterisation of a mixture from a component-wise decomposition to a construct made of the first moment(s) of the distribution and of component-wise objects constrained by the moment equation(s). There is of course a bijection between both parameterisations, but the constraints appearing in the latter produce compact parameter spaces for which (different) uniform priors can be proposed. While the resulting posteriors are no longer conjugate, even conditional on the latent variables, standard Metropolis algorithms can be implemented to produce Monte Carlo approximations of these posteriors.
[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.
Today, I alas missed a seminar at BiPS on the Zig-Zag (sub-)sampler of Joris Bierkens, Paul Fearnhead and Gareth Roberts, presented here in Paris by James Ridgway. Fortunately for me, I had some discussions with Murray Pollock in Warwick and then again with Changye Wu in Dauphine that shed some light on this complex but highly innovative approach to simulating in Big Data settings thanks to a correct subsampling mechanism.
The zig-zag process runs a continuous process made of segments that turn from one diagonal to the next at random times driven by a generator connected with the components of the gradient of the target log-density. Plus a symmetric term. Provided those random times can be generated, this process is truly available and associated with the right target distribution. When the components of the parameter are independent (an unlikely setting), those random times can be associated with an inhomogeneous Poisson process. In the general case, one needs to bound the gradients by more manageable functions that create a Poisson process that can later be thinned. Next, one needs to simulate the process for the upper bound, a task that seems hard to achieve apart from linear and piecewise constant upper bounds. The process has a bit of a slice sampling taste, except that it cannot be used as a slice sampler but requires continuous time integration, given that the length of each segment matters. (Or maybe random time subsampling?)
A highly innovative part of the paper concentrates on Big Data likelihoods and on the possibility to subsample properly and exactly the original dataset. The authors propose Zig-Zag with subsampling by turning the gradients into random parts of the gradients. While remaining unbiased. There may be a cost associated with this gain of one to n, namely that the upper bounds may turn larger as they handle all elements in the likelihood at once, hence become (even) less efficient. (I am more uncertain about the case of the control variates, as it relies on a Lipschitz assumption.) While I still miss an easy way to implement the approach in a specific model, I remain hopeful for this new approach to make a major dent in the current methodologies!
“An anytime algorithm is an algorithm that can be run continuously, generating progressively better solutions when afforded additional computation time. Traditional particle-based inference algorithms are not anytime in nature; all particles need to be propagated in lock-step to completion in order to compute expectations.”
Following a discussion with Lawrence Murray last week, I read Paige et al. NIPS 2014 paper on their anytime sequential Monte Carlo algorithm. As explained above, an anytime algorithm is interruptible, meaning it can be stopped at any time without biasing the outcome of the algorithm. While MCMC algorithms can qualify as anytime (provided they are in stationary regime), it is not the case with sequential and particle Monte Carlo algorithms, which do not have an inbred growing mechanism preserving the target. In the case of Paige et al.’s proposal, the interruptible solution returns an unbiased estimator of the marginal likelihood at time n for any number of particles, even when this number is set or increased during the computation. The idea behind the solution is to create a particle cascade by going one particle at a time and creating children of this particle in proportion to the current average weight. An approach that can be run indefinitely. And since memory is not infinite, the authors explain how to cap the number of alive particles without putting the running distribution in jeopardy…