11am: Fast Dynamic Posterior Exploration for Factor Augmented Multivariate Regression byVeronika Rockova
Advancements in high-throughput experimental techniques have facilitated the availability of diverse genomic data, which provide complementary information regarding the function and organization of gene regulatory mechanisms. The massive accumulation of data has increased demands for more elaborate modeling approaches that combine the multiple data platforms. We consider a sparse factor regression model, which augments the multivariate regression approach by adding a latent factor structure, thereby allowing for dependent patterns of marginal covariance between the responses. In order to enable the identication of parsimonious structure, we impose spike and slab priors on the individual entries in the factor loading and regression matrices. The continuous relaxation of the point mass spike and slab enables the implementation of a rapid EM inferential procedure for dynamic posterior model exploration. This is accomplished by considering a nested sequence of spike and slab priors and various factor space cardinalities. Identied candidate models are evaluated by a conditional posterior model probability criterion, permitting trans-dimensional comparisons. Patterned sparsity manifestations such as an orthogonal allocation of zeros in factor loadings are facilitated by structured priors on the binary inclusion matrix. The model is applied to a problem of integrating two genomic datasets, where expression of microRNA’s is related to the expression of genes with an underlying connectivity pathway network.
Archive for g-prior
Following Ed George (Wharton) and Feng Liang (University of Illinois at Urbana-Champaign) talks today in Dauphine, Natalia Bochkina (University of Edinburgh) will give a talk on Thursday, June 20, at 2pm in Room 18 at ENSAE (Malakoff) [not Dauphine!]. Here is her abstract:
2 am: Simultaneous local and global adaptivity of Bayesian wavelet estimators in nonparametric regression by Natalia Bochkina
We consider wavelet estimators in the context of nonparametric regression, with the aim of finding estimators that simultaneously achieve the local and global adaptive minimax rate of convergence. It is known that one estimator – James-Stein block thresholding estimator of T.Cai (2008) – achieves simultaneously both optimal rates of convergence but over a limited set of Besov spaces; in particular, over the sets of spatially inhomogeneous functions (with 1≤ p<2) the upper bound on the global rate of this estimator is slower than the optimal minimax rate.
Another possible candidate to achieve both rates of convergence simultaneously is the Empirical Bayes estimator of Johnstone and Silverman (2005) which is an adaptive estimator that achieves the global minimax rate over a wide rage of Besov spaces and Besov balls. The maximum marginal likelihood approach is used to estimate the hyperparameters, and it can be interpreted as a Bayesian estimator with a uniform prior. We show that it also achieves the adaptive local minimax rate over all Besov spaces, and hence it does indeed achieve both local and global rates of convergence simultaneously over Besov spaces. We also give an example of how it works in practice.
Next week, we are having a special Bayesian [top] model choice week in Dauphine, thanks to the simultaneous visits of Ed George (Wharton), Feng Liang (University of Illinois at Urbana-Champaign), and Veronika Rockovà (Erasmus University). To start the week and get to know the local actors (!), Ed and Feng both give a talk on Tuesday, June 18, at 11am and 1pm in Room C108. Here are the abstracts:
11am: Prediction and Model Selection for Multi-task Learning by Feng Liang
In multi-task learning one simultaneoulsy fits multiple regression models. We are interested in inference problems like model selection and prediction when there are a large number of tasks. A simple version of such models is a one-way ANOVA model where the number of replicates is fixed but the number of groups goes to infinity. We examine the consistency of Bayesian procedures using Zellner (1986)’s g-prior and its variants (such as mixed g-priors and Empirical Bayes), and compare their prediction accuracy with other procedures, such as the ones based AIC/BIC and group Lasso. Our results indicate that the Empirical Bayes procedure (with some modification for the large p small n setting) can achieve model selection consistency, and also have better estimation accuracy than other procedures being considered. During my talk, I’ll focus on the analysis on the one-way ANOVA model, but will also give a summary on our findings for multi-tasking learning invovling a more general regression setting. This is based on joint work with my PhD student Bin Li from University of Illinois at Urbana-Champaign.
1pm: EMVS: The EM Approach to Bayesian Variable Selection by Edward George
Despite rapid developments in stochastic search algorithms, the practicality of Bayesian variable selection methods has continued to pose challenges. High-dimensional data are now routinely analyzed, typically with many more covariates than observations. To broaden the applicability of Bayesian variable selection for such high-dimensional linear regression contexts, we propose EMVS, a deterministic alternative to stochastic search based on an EM algorithm which exploits a conjugate mixture prior formulation to quickly find posterior modes. Combining a spike-and-slab regularization diagram for the discovery of active predictor sets with subsequent rigorous evaluation of posterior model probabilities, EMVS rapidly identifies promising sparse high posterior probability submodels. External structural information such as likely covariate groupings or network topologies is easily incorporated into the EMVS framework. Deterministic annealing variants are seen to improve the effectiveness of our algorithms by mitigating the posterior multi-modality associated with variable selection priors. The usefulness the EMVS approach is demonstrated on real high-dimensional data, where computational complexity renders stochastic search to be less practical. This is joint work with Veronika Rockova of Erasmus University)
Another day full of interesting and challenging—in the sense they generated new questions for me—talks at the SuSTain workshop. After another (dry and fast) run around the Downs; Leo Held started the talks with one of my favourite topics, namely the theory of g-priors in generalized linear models. He did bring a new perspective on the subject, introducing the notion of a testing Bayes factor based on the residual statistic produced by a classical (maximum likelihood) analysis, connected with earlier works of Vale Johnson. While I did not truly get the motivation for switching from the original data to this less informative quantity, I find this perspective opening new questions for dealing with settings where the true data is replaced with one or several classical statistics. With possible strong connections to ABC, of course. Incidentally, Leo managed to produce a napkin with Peter Green’s intro to MCMC dating back from their first meeting in 1994: a feat I certainly could not reproduce (as I also met both Peter and Leo for the first time in 1994, at CIRM)… Then Richard Everit presented his recent JCGS paper on Bayesian inference on latent Markov random fields, centred on the issue that simulating the latent MRF involves an MCMC step that is not exact (as in our earlier ABC paper for Ising models with Aude Grelaud). I already discussed this paper in an earlier blog and the only additional question that comes to my mind is whether or not a comparison with the auxiliary variable approach of Møller et al. (2006) would make sense.
In the intermission, I had a great conversation with Oliver Ratman on his talk of yesterday on the surprising feature that some models produce as “data” some sample from a pseudo-posterior.. Opening once again new vistas! The following talks were more on the mathematical side, with James Cussens focussing on the use of integer programming for Bayesian variable selections, then Éric Moulines presenting a recent work with a PhD student of his on PAC-Bayesian bounds and the superiority of combining experts. Including a CRAN package. Éric concluded his talk with the funny occurence of Peter’s photograph on Éric’s Microsoft Research Profile own page, due to Éric posting our joint photograph at the top of Pic du Midi d’Ossau in 2005… (He concluded with a picture of the mountain that was the exact symmetry of mine yesterday!)
The afternoon was equally superb with Gareth Roberts covering fifteen years of scaling MCMC algorithms, from the mythical 0.234 figure to the optimal temperature decrease in simulated annealing, John Kent playing the outlier with an EM algorithm—however including a formal prior distribution and raising the challenge as to why Bayesians never had to constrain the posterior expectation, which prompted me to infer that (a) the prior distribution should include all constraints and (b) the posterior expectation was not the “right” tool in non-convex parameters spaces—. Natalia Bochkina presented a recent work, joint with Peter Green, on connecting image analysis with Bayesian asymptotics, reminding me of my early attempts at reading Ibragimov and Has’minskii in the 1990’s. Then a second work with Vladimir Spoikoini on Bayesian asymptotics with misspecified models, introducing a new notion of effective dimension. The last talk of the day was by Nils Hjort about his coming book on “Credibility, confidence and likelihood“—not yet advertised by CUP—which sounds like an attempt at resuscitating Fisher by deriving distributions in the parameter space from frequentist confidence intervals. I already discussed this notion in an earlier blog, so I am fairly skeptical about it, but the talk was representative of Nils’ highly entertaining and though-provoking style! Esp. as he sprinkled the talk with examples where MLE (and some default Bayes estimators) did not work. And reanalysed one of Chris Sims‘ example presented during his Nobel Prize talk…
After a huge delay, since the project started in 2006 and was first presented in Banff in 2007 (as well as included in the Bayesian Core), Gilles Celeux, Mohammed El Anbari, Jean-Michel Marin, and myself have eventually completed our paper on using hyper-g priors variable selection and regularisation in linear models . The redaction of this paper was mostly delayed due to the publication of the 2007 JASA paper by Feng Liang, Rui Paulo, German Molina, Jim Berger, and Merlise Clyde, Mixtures of g-priors for Bayesian variable selection. We had indeed (independently) obtained very similar derivations based on hypergeometric function representations but, once the above paper was published, we needed to add material to our derivation and chose to run a comparison study between Bayesian and non-Bayesian methods for a series of simulated and true examples. It took a while to Mohammed El Anbari to complete this simulation study and even longer for the four of us to convene and agree on the presentation of the paper. The only difference between Liang et al.’s (2007) modelling and ours is that we do not distinguish between the intercept and the other regression coefficients in the linear model. On the one hand, this gives us one degree of freedom that allows us to pick an improper prior on the variance parameter. On the other hand, our posterior distribution is not invariant under location transforms, which was a point we heavily debated in Banff… The simulation part shows that all “standard” Bayesian solutions lead to very similar decisions and that they are much more parsimonious than regularisation techniques.
Two other papers posted on arXiv today address the model choice issue. The first one by Bruce Lindsay and Jiawei Liu introduces a credibility index, and the second one by Bazerque, Mateos, and Giannakis considers group-lasso on splines for spectrum cartography.
Earlier this month, Daniel Sabanés Bové and Leo Held posted a paper about g-priors on arXiv. While I glanced at it for a few minutes, I did not have the chance to get a proper look at it till last Sunday. The g-prior was first introduced by the late Arnold Zellner for (standard) linear models, but they can be extended to generalised linear models (formalised by the late John Nelder) at little cost. In Bayesian Core, Jean-Michel Marin and I do centre the prior modelling in both linear and generalised linear models around g-priors, using the naïve extension for generalised linear models,
as in the linear case. Indeed, the reasonable alternative would be to include the true information matrix but since it depends on the parameter outside the normal case this is not truly an alternative. Bové and Held propose a slightly different version
where W is a diagonal weight matrix and c is a family dependent scale factor evaluated at the mode 0. As in Liang et al. (2008, JASA) and most of the current literature, they also separate the intercept from the other regression coefficients. They also burn their “improperness joker” by choosing a flat prior on , which means they need to use a proper prior on g, again as Liang et al. (2008, JASA), for the corresponding Bayesian model comparison to be valid. In Bayesian Core, we do not separate from the other regression coefficients and hence are left with one degree of freedom that we spend in choosing an improper prior on g instead. (Hence I do not get the remark of Bové and Held that our choice “prohibits Bayes factor comparisons with the null model“. As argued in Bayesian Core, the factor g being an hyperparameter shared by all models, we can use the same improper prior on g in all models and hence use standard Bayes factors.) In order to achieve closed form expressions, the authors use Cui and George ‘s (2008) prior
which requires the two hyper-hyper-parameters a and b to be specified.
The second part of the paper considers computational issues. It compares the ILA solution of Rue, Martino and Chopin (2009, Series B) with an MCMC solution based on an independent proposal on g resulting from linear interpolations (?). The marginal likelihoods are approximated by Chib and Jeliazkov (2001, JASA) for the MCMC part. Unsurprisingly, ILA does much better, even with a 97% acceptance rate in the MCMC algorithm.
The paper is very well-written and quite informative about the existing literature. It also uses the Pima Indian dataset (The authors even dug out a 1991 paper of mine I had completely forgotten!) I am actually thinking of using the review in our revision of Bayesian Core, even though I think we should stick to our choice of including within the set of parameters…