Bayesian News Feeds

OxWaSP (The Oxford-Warwick Statistics Programme)

Xian's Og - Tue, 2014-01-21 09:18

This is an official email promoting OxWaSP, our joint doctoral training programme, which I [objectively] think is definitely worth considering if planning a PhD in Statistics. Anywhere.

The Statistics Department – University of Oxford and the Statistics Department – University Of Warwick, supported by the EPSRC, will run a joint Centre of Doctoral Training in the theory, methods and applications of Statistical Science for 21st Century data-intensive environments and large-scale models. This is the first centre of its type in the World and will equip its students to work in an area in growing demand both in academia and industry.

Each year from October 2014 OxWaSP will recruit at least 5 students attached to Warwick and at least 5 attached to Oxford. Each student will be funded with a grant for four years of study. Students spend the first year at Oxford developing advanced skills in statistical science. In the first two terms students are given research training through modular courses: Statistical Inference in Complex Models; Multivariate Stochastic Processes; Bayesian Analyses for Complex Structural Information; Machine Learning and Probabilistic Graphical Models; Stochastic Computation for Intractable Inference. In the third term, students carry out two small research projects. At the end of year 1, students begin a three-year research project with a chosen supervisor, five continuing at Oxford and five moving to the University of Warwick.

Training in years 2-4 includes annual retreats, workshops and a research course in machine learning at Amazon (Berlin). There are funded opportunities for students to work with our leading industrial partners and to travel in their third year to an international summer placement in some of the strongest Statistics groups in the USA, Europe and Asia including UC Berkeley, Columbia University, Duke University, the University of Washington in Seattle, ETH Zurich and NUS Singapore.

Applications will be considered in gathered fields with the next deadline of 24 January 2014 (Non-EU applicants should apply by this date to maximise their chance of funding). Interviews for successful applicants who submit by the January deadline will take place at the end of February 2014. There will be a second deadline for applications at the end of February (Warwick) and 14th March (Oxford).


Filed under: Kids, Statistics, University life Tagged: big data, data science, EPSRC, machine learning, PhD position, University of Oxford, University of Warwick

Categories: Bayesian Bloggers

Bayesian inference for low count time series models with intractable likelihoods

Xian's Og - Mon, 2014-01-20 19:14

Last evening, I read a nice paper with the above title by Drovandi, Pettitt and McCutchan, from QUT, Brisbane. Low count refers to observation with a small number of integer values. The idea is to mix ABC with the unbiased estimators of the likelihood proposed by Andrieu and Roberts (2009) and with particle MCMC… And even with a RJMCMC version. The special feature that makes the proposal work is that the low count features allows for a simulation of pseudo-observations (and auxiliary variables) that may sometimes authorise an exact constraint (that the simulated observation equals the true observation). And which otherwise borrows from Jasra et al. (2013) “alive particle” trick that turns a negative binomial draw into an unbiased estimation of the ABC target… The current paper helped me realise how powerful this trick is. (The original paper was arXived at a time I was off, so I completely missed it…) The examples studied in the paper may sound a wee bit formal, but they could lead to a better understanding of the method since alternatives could be available (?). Note that all those examples are not ABC per se in that the tolerance is always equal to zero.

The paper also includes reversible jump implementations. While it is interesting to see that ABC (in the authors’ sense) can be mixed with RJMCMC, it is delicate to get a feeling about the precision of the results, without a benchmark to compare to. I am also wondering about less costly alternatives like empirical likelihood and other ABC alternatives. Since Chris is visiting Warwick at the moment, I am sure we can discuss this issue next week there.


Filed under: Books, Statistics, Travel, University life Tagged: ABC, Brisbane, discrete time series models, hidden Markov models, particle filter, particle MCMC, QUT, RJMCMC, University of Warwick
Categories: Bayesian Bloggers

Postup Donja Banda

Xian's Og - Mon, 2014-01-20 15:20


Filed under: Wines Tagged: Croatia, Plavac Mali grape, Postup wine
Categories: Bayesian Bloggers

The Missing Picture nominated for the Oscars!

Xian's Og - Sun, 2014-01-19 19:13

The Missing Picture (L’image manquante), a documentary movie by Rithy Panh whose text is read by my friend, co-author and former student Randal Douc, has just been nominated for the Oscars. (After winning a prize in Cannes.) As a foreign-language movie. This film is actually much much more than a documentary on the Red Khmer terror. The text is just gripping and the idea of depicting the horror of the camps by little clay figures made and painted in the movie gives them a strength and a life other representations would miss. The painstaking application of the sculptor into making those figures endows them with a soul that make their suffering the more real. The Missing Picture goes beyond the political and the ideological to translate the simple but absolute multiplication of lost destinies. Whether or not it stands a chance for the Oscars, a truly unique movie.


Filed under: Books, pictures Tagged: Cambodgia, Oscars (Academy Awards), Randal Douc, Red Khmers, Rithy Panh, The Missing Picture
Categories: Bayesian Bloggers

Le Monde puzzle [#849]

Xian's Og - Sat, 2014-01-18 19:14

A straightforward Le Monde mathematical puzzle:

Find a pair (a,b) of integers such that a has an odd number d of digits larger than 2 and ab is written as 10d+1+10a+1. Find the smallest possible values of a and of b.

I ran the following R code

d=3 for (a in 10^(d-1):(10^d-1)){ c=10^(d+1)+10*a+1 if (a*trunc(c/a)==c) print(c(a,c))}

which produced a=137 (and b=83) as the unique case. For d=4, I obtained a=9091 and b=21, for d=6, a=909091, and b=21, for d=7, a=5882353 and b=27, while for d=5, my code did not return any solution. While d=8 took too long to run, a prime factor decomposition of 10⁹+1 leads to (with the schoolmath R library)

> for (d in 3:10) print(c(d,prime.factor(10^(d+1)+1))) [1] 3 73 137 [1] 4 11 9091 [1] 5 101 9901 [1] 6 11 909091 [1] 7 17 5882353 [1] 8 7 11 13 19 52579 [1] 9 101 3541 27961 [1] 10 11 11 23 4093 8779

which gives a=52631579 and b=29 for d=8 and also explains why there is no solution for d=5. The corresponding a has too many digits!

This issue of Le Monde Science&Médecine leaflet had more interesting entries, from one on “LaTeX as the lingua franca of mathematicians”—which presumably made little sense to any reader unfamiliar with LaTeX—to the use of “big data” tools (like news rover) to analyse data produce by the medias, to  yet another tribune of Marco Zito about the “five sigma” rule used in particle physics (and for the Higgs boson analysis)—with the reasonable comment that a large number of repetitions of an experiment is likely to exhibit unlikely events, and an also reasonable recommendation to support “reproduction experiments” that aim at repeating exceptional phenomena—, to a solution to puzzle #848—where the resolution is the same as mine’s, but mentions the principle of Dirichlet’s drawers to exclude the fact that all prices are different, a principle I had never heard off…


Filed under: Books, Kids, R, Statistics Tagged: five sigma, Higgs boson, LaTeX, Le Monde, mathematical puzzle, particle physics
Categories: Bayesian Bloggers

no Complaints [book review]

Xian's Og - Fri, 2014-01-17 19:14

Another Rankin! In the Complaints series: the main character is Malcom Fox, inspector at the “Complaints and Conduct Department”, investigating a case of corruption within the Force on the north side of the Forth, in Fife. Rankin builds on the history of Scottish violent nationalist groups in the 80′s to deliver a very convincing story, mixing as usual the unorthodox methods of an investigator with his personal life. Even though some tie-ins are a wee bit unrealistic and I do not buy the final  (major) scene, I enjoyed reading the book over two or three days (between Chamonix, Geneva and Paris). Maybe due to the novelty of the character, there is no feeling of repetitiveness in this instalment. And the background is definitely interesting, relating the older SNP with violent splint groups at a time when Scottish independence was beyond the realm of the possible. I am now looking forward the next instalment, Saints of the Shadow Bible, where Fox and Rebus share the scene….


Filed under: Books Tagged: Edinburgh, Fife, Ian Rankin, Rebus, Scotland, SNP
Categories: Bayesian Bloggers

robust Bayesian FDR control with Bayes factors [a reply]

Xian's Og - Thu, 2014-01-16 19:14

(Following my earlier discussion of his paper, Xiaoquan Wen sent me this detailed reply.)

I think it is appropriate to start my response to your comments by introducing a little bit of the background information on my research interest and the project itself: I consider myself as an applied statistician, not a theorist, and I am interested in developing theoretically sound and computationally efficient methods to solve practical problems. The FDR project originated from a practical application in genomics involving hypothesis testing. The details of this particular application can be found in this published paper, and the simulations in the manuscript are also designed for a similar context. In this application, the null model is trivially defined, however there exist finitely many alternative scenarios for each test. We proposed a Bayesian solution that handles this complex setting quite nicely: in brief, we chose to model each possible alternative scenario parametrically, and by taking advantage of Bayesian model averaging, Bayes factor naturally ended up as our test statistic. We had no problem in demonstrating the resulting Bayes factor is much more powerful than the existing approaches, even accounting for the prior (mis-)modeling for Bayes factors. However, in this genomics application, there are potentially tens of thousands of tests need to be simultaneously performed, and FDR control becomes necessary and challenging.

We actually started with a full Bayesian approach following the framework proposed by Newton et al (2004) by treating multiple testing as a mixture of null and (many components of) alternative models, and intrinsically making inference w.r.t the proportion of true nulls (I will call this quantity pi0 from this point on). The difficulty is related to the inference of pi0: although for controlling FDR, we were only interested in identifying two classes (null and non-null) from the mixture, the mixture itself can have many more components induced by different alternative models. Through simulations, we observed that the inference of pi0, e.g. its credible interval, is highly sensitive to its prior specification in the mixture model: e.g. the uniform prior led to severe underestimating of pi0 and consequently inflated FDR. (It should be noted, in the simulations, the assumed alternative models and data generative alternative models are reasonably similar, and we also explicitly model the many components of alternatives.) Although applying a more sophisticated Bayesian inference framework (e.g. Dirichlet process type of prior) might resolve this issue, we did not pursue this direction, largely because of the concerns of computational burdens on this large scale data problem. Instead, we turned to the famous Bayes/Non-Bayes compromise, i.e., treating Bayes factor as a test statistic and applying permutations to find its p-value and apply p-value based FDR control procedure. We were actually satisfied with this approach to certain degree and published our application paper using this solution. Until recently, when we started analyzing a scaled-up data set, we realized the permutation p-value scheme hit some computational bottleneck and became impractical. This motivated me to re-think a computational efficient solution to control FDR using Bayes factors.

By telling this story, I’d hope that you would agree with me on the following points:

1. We care very much about the alternative hypothesis and that is why we choose Bayes factors (which require explicit parametric modeling of the alternatives) in the first place. The first quote from the manuscript in your comments,

“Although the Bayesian FDR control is conceptually straightforward, its practical performance is susceptible to alternative model misspecifications. In comparison, the p-value based frequentist FDR control procedures demand only adequate behavior of p-values under the null models and generally ensure targeted FDR control levels, regardless of the distribution of p-values under the assumed alternatives.”

is a critical one, but my emphasis here is that “Bayesian FDR control procedure” is susceptible to “alternative model misspecification”, not so much on Bayes factor itself. I think it boils down to the point if one thinks controlling pre-defined FDR level is a meaningful thing. For us, to compare our Bayes factor solution with the existing frequentist approaches in the application mentioned above, we had to value this standard. Finally, from a mathematical point of view, this can be achieved with Bayes factors in principle, as illustrated in the Bayes/Non-Bayes compromise.

2. I think the mixture model formulation of multiple hypothesis testing problem is quite common in both frequentist and Bayesian practice. But my observation is that the main difficulty seems to be making accurate inference of pi0 when alternative models are highly heterogeneous. In simulation 1 of the manuscript, we made a comparison with the localFDR procedure where the alternatives are inferred by a spline Poisson regression density estimation method with default parameters, and in cases, we found pi0 severely under-estimated (and fdr inflated) if care is not taken. I believe that sophisticated Bayesian/parametric solutions might resolve this issue, but most of the things I know do not scale up computationally to the practical problem that we cared about.

With these explained, I am now ready to present my interpretation of the proposed Bayesian FDR procedure: the proposed procedure attempts to guard pre-defined FDR level rigorously in a computational efficient way, even when the alternatives are highly heterogeneous and/or “accurate” parametric specification is difficult to find (as a result, the accurate inference of pi0 is computationally non-trivial). The proposed procedure has at least two major advantages over the Bayes/Non-Bayes compromise solution:

1. The computational efficiency: the EBF procedure described in the paper does not need any permutation; the QBF procedure may need permutations to estimate the median of Bayes factor distribution under the null but required permutations would be much less than the numbers required to accurately estimate p-values.

2. The procedure is actually controlling for Bayesian FDR not Frequentist FDR, which might be a theoretical advantage and also differs from B/N-B compromise.

Furthermore, I’d like interpret the proposed procedure a computational approximation (albeit a conservative one) to the true Bayesian FDR control procedure described in Newton et al (2004), Mueller et al (2004), and Mueller et al (2006), and therefore purely Bayesian, philosophically speaking. To see this, it is critical to interpret hat-pi0 (equations 3.1 to 3.3 on page 9 in the manuscript) as a probability upper bound of the posterior distribution of pi0 regardless of its prior specification. The “regardless” part comes from an argument of Bayesian asymptotics when number of the simultaneous tests is large (explained also on page 9), which is not an unreasonable assumption given the applications we are facing everyday in genomics. This also should provide intuitions for the way we estimate hat-pi0 by simply applying the law of large numbers (LLN). I agree with your assessment that the procedure looks frequentist-y, but at the same time, I don’t believe that LLN is a patent of frequentist and the part of the procedure should be put into the big picture of the overall scheme. I also acknowledge that this procedure is conservative, but in a similar level as Storey’s procedure and generally better than the Benjamini-Hochberg procedure.

I have to say that I am a bit surprised that you selected the second quote as a representative summary from the manuscript — it really is not. It only aimed to provide a little intuition that sample mean of the BFs carries information about pi0, nothing further. If I could pick a summarizing quote from the manuscript, I’d like choose the following one from the discussion section:

“We have introduced a Bayesian FDR control procedure with Bayes factors that is robust to misspecifications of alternative models. This feature should provide peace of mind for practitioners who are attempting parametric Bayesian models in multiple hypothesis testing. Nevertheless, within our framework, the model specification still dictates the overall performance, e.g., a badly designed alternative model would have very little power and would therefore be useless. Our central message throughout this paper has been that various FDR control procedures have little practical difference if the same or similar test statistics are applied; however, our proposed procedure encourages well-designed parametric modeling approaches to obtain more powerful test statistics.”

Finally, on your final comments that the distribution of Bayes factors can be useful to calibrate approximations, I completely agree! As a matter of fact, in a paper currently in press at Annals of Applied Statistics, Matthew Stephens and I have utilized the exact idea to calibrate the approximate Bayes factors computed from the Laplace approximation for finite sample size situations, and it worked amazingly well (see appendix D)!


Filed under: Statistics, University life Tagged: Bayes factors, design of experiments, Dirichlet process, empirical Bayes methods, FDRs
Categories: Bayesian Bloggers

MCMSki 4, 5… [rejuvenating suggestion]

Xian's Og - Wed, 2014-01-15 19:14

Another thing I should have included in the program. Or in the organising committee: a link with the Young Bayesians (j-ISBA) section… As pointed out to me by Kerrie Mengersen, ISBA meetings are obvious opportunities for young researchers to interact and network, as well as for seeking a job. Thus, there should be time slots dedicated to them in every ISBA sponsored meeting, from a mixer on the very first day to a job market coffee break the next day (and to any other social activity bound to increase the interactivity. Like a ski race.). So I would suggest every ISBA sponsored event (and no only the Bayesian Young Statistician Meetings!) should include a j-ISBA representative in its committee(s) to enforce this policy… (Kerrie also suggested random permutations during the banquet which is a neat idea provided the restaurant structure allows for this. It would have been total chaos in La Calèche last week!)


Filed under: Kids, Mountains, pictures, Statistics, Travel, University life Tagged: BAYSM 2014, Chamonix, j-ISBA, job market, MCMSki IV, Wien
Categories: Bayesian Bloggers

M(CM)Ski 5? [with ski polls]

Xian's Og - Tue, 2014-01-14 19:14

Along other members of BayesComp who launched a brainstorming session for the next MCMSki meeting before the snow has completely melted from our skis, we discussed the following topics about the future meeting:

1. Should we keep the brandname MCMSki for the incoming meetings? The argument for changing the name is that the community is broader than MCMC, as already shown by the program of MCMCSki 4.  I have no strong feeling about this name, even though I find it catchy and sexy! I would thus rather keep MCMSki because it is already a brandname. Else, we could switch to M(CM)Ski, MCMSki with friends (and foes?), Snowtistics and Compuskis, or to any other short name with or without ski in it, as long as the filiation from the previous meetings is clear in the mind of the participants. Take Our Poll (function(d,c,j){if(!d.getElementById(j)){var pd=d.createElement(c),s;pd.id=j;pd.src='https://s1.wp.com/wp-content/mu-plugins/shortcodes/js/polldaddy-shortcode.js';s=d.getElementsByTagName(c)[0];s.parentNode.insertBefore(pd,s);} else if(typeof jQuery !=='undefined')jQuery(d.body).trigger('pd-script-load');}(document,'script','pd-polldaddy-loader'));

2. Should we move the frequency to two years? While the current meeting was highly popular and attracted the record number of 223 participants, and while the period right after the Winter break is not so heavily packed with meetings, we were several at a banquet table last week to object to a planned move from three to two years. I understand the appeal of meetings with great speakers in a terrific mountainous taking place as often as possible… However what stroke me with the meeting last week is that, despite the large number of parallel sessions, I overwhelmingly heard novel stuff, compared with previous meetings. And would have heard even more, had I been gifted with ubiquity. Moving to two years could cull this feeling. And induce “meeting fatigue. Furthermore, I fear that the increase in ISBA sections and the natural increase of meeting entropy pushes the percentage of meetings one can attend down and down. Sticking to a three year period would keep MCMSki more significantly attractive in that refusing an invitation would mean postponing for three years, &tc. So I personally oppose a move to two years. Take Our Poll (function(d,c,j){if(!d.getElementById(j)){var pd=d.createElement(c),s;pd.id=j;pd.src='https://s1.wp.com/wp-content/mu-plugins/shortcodes/js/polldaddy-shortcode.js';s=d.getElementsByTagName(c)[0];s.parentNode.insertBefore(pd,s);} else if(typeof jQuery !=='undefined')jQuery(d.body).trigger('pd-script-load');}(document,'script','pd-polldaddy-loader'));

3. Should we seek further financial support? The financial support behind a conference is obviously crucial. When planning MCMski 4, I however decided against contacting companies as I have no skills in the matter, but finding ways to support conference rooms, youngster travels, ski race, poster prizes and banquet would be more-than-nice. Anto’s initiative to bring a pair of skis offered by a ski company was a great one and one feat that I hope can be duplicated in the future. (During my spare week in Chamonix, I contacted ski rentals and the skipass company for a rebate, to no avail.) Travel support from ISBA and SBSS towards the travel costs of around 20 young researchers was much appreciated but is not necessarily to be found at each occurrence… Note that, despite the lack of corporate support, MCMski 4 is going to provide a positive financial return to ISBA (and BayesComp) and I strongly suggest we keep a tradition of minimalist services for the future meetings in order to fight outrageous conference fees. I think the fees should cover the conference rooms and possibly a cuppa or two a day but nothing more. In particular, the banquet should remain optional. And so should any other paying social event. (We can also do without goodies and conference material.)

4. Where should the next meeting take place? The call is on for potential organisers in either 2016 or 2017, early January. Between the Alps and the Rockies, there are plenty of possible locations, but more exotic places in the Northern Hemisphere could be suggested as well, from Lapland to Hokkaido… A question raised by Christophe Andrieu that I’d like to second is whether the preference should go to places that qualify as villages or as resort. Bormio and Chamonix are villages, while Park City is not. (I definitely prefer villages!) Take Our Poll (function(d,c,j){if(!d.getElementById(j)){var pd=d.createElement(c),s;pd.id=j;pd.src='https://s1.wp.com/wp-content/mu-plugins/shortcodes/js/polldaddy-shortcode.js';s=d.getElementsByTagName(c)[0];s.parentNode.insertBefore(pd,s);} else if(typeof jQuery !=='undefined')jQuery(d.body).trigger('pd-script-load');}(document,'script','pd-polldaddy-loader'));


Filed under: Mountains, pictures, Statistics, Travel, University life Tagged: BayesComp, Chamonix-Mont-Blanc, ISBA conference, MCMSki
Categories: Bayesian Bloggers

accelerated ABC

Xian's Og - Mon, 2014-01-13 19:14

Richard Wilkinson arXived a paper on accelerated ABC during MCMSki 4, paper that I almost missed when quickly perusing the daily list. This is another illustration of the “invasion of Gaussian processes” in ABC settings. Maybe under the influence of machine learning.

The paper starts with a link to the synthetic likelihood approximation of Wood (2010, Nature), as in Richard Everitt’s talk last week. Richard (W.) presents the generalised ABC as a kernel-based acceptance probability, using a kernel π(y|x), when y is the observed data and x=x(θ) the simulated one. He proposes a Gaussian process modelling for the log-likelihood (at the observed data y), with a quadratic (in θ) mean and Matérn covariance matrix. Hence the connection with Wood’s synthetic likelihood. Another connection is with Nicolas’ talk on QMC(MC): the θ’s are chosen following a Sobol sequence “in order to minimize the number of design points”. Which requires a reparameterisation to [0,1]p… I find this “uniform” exploration of the whole parameter space delicate to envision in complex parameter spaces and realistic problems, since the likelihood is highly concentrated on a tiny subregion of the original [0,1]p. Not mentioning the issue of the spurious mass on the boundaries of the hypercube possibly induced by the change of variable. The sequential algorithm of Richard also attempts at eliminating implausible zones of the parameter space. i.e. zones where the likelihood is essentially zero. My worries with this interesting notion are that (a) the early Gaussian process approximations may be poor and hence exclude zones they should not; (b) all Gaussian process approximations at all iterations must be saved; (c) the Sobol sequences apply to the whole [0,1]p at each iteration but the non-implausible region shrinks at each iteration, which induces a growing inefficiency in the algorithm. The Sobol sequence should be restricted to the previous non-implausible zone.

Overall, an interesting proposal that would need more prodding to understand whether or not it is robust to poor initialisation and complex structures. And a proposal belonging to the estimated likelihood branch of ABC, which makes use of the final Gaussian process approximation to run an MCM algorithm. Without returning to pseudo-data simulation, replacing it with log-likelihood simulation.

“These algorithms sample space randomly and naively and do not learn from previous simulations”

The above criticism is moderated in a footnote about ABC-SMC using the “current parameter value to determine which move to make next [but] parameters visited in previous iterations are not taken into account”. I still find it excessive in that SMC algorithms and in particular ABC-SMC algorithms are completely free to use the whole past to build the new proposal. This was clearly enunciated in our earlier population Monte Carlo papers. For instance, the complete collection of past particles can be recycled by weights computing thru our AMIS algorithm, as illustrated by Jukka Corander in one genetics application.


Filed under: Books, Mountains, Statistics Tagged: ABC, accelerated ABC, Chamonix-Mont-Blanc, Gaussian processes, MCMSki IV, MCQMC, Sobol sequences
Categories: Bayesian Bloggers

MCMSki IV [mistakes and regrets]

Xian's Og - Sun, 2014-01-12 19:14

Now that the conference and the Bayesian non-parametric satellite workshop (thanks to Judith!) are over, with (almost) everyone back home, and that the post-partum conference blues settles in (!), I can reflect on how things ran for those meetings and what I could have done to improve them… (Not yet considering to propose a second edition of MCMSki in Chamonix, obviously!)

Although this was clearly a side issue for most participants, the fact that the ski race did not take place still rattles me!  In retrospect, adding a mere 5€ amount to the registration fees for all participants would have been enough to cover the (fairly high) fares asked by the local ski school. Late planning for the ski race led to overlook this basic fact…

Since MCMSki is now the official conference of the BayesComp section of ISBA, I should have planned well in advance a section meeting within the program, if only to discuss the structure of the next meeting and how to keep the section alive. Waiting till the end of the last section of the final day was not the best idea!

Another thing I postponed for too long was seeking some sponsors: fortunately, the O’Bayes meeting in Duke woke me up to the potential of a poster prize and re-fortunately Academic Press, CRC Press, and Springer-Verlag reacted quickly enough to have plenty of books to hand to the winners. If we could have had another group of sponsors financing a beanie or something similar, it would have been an additional perk… Even though I gathered enough support from participants about the minimalist conference “package” made of a single A4 sheet.

Last, I did not advertise properly on the webpage and at all during the meeting for the special issue of Statistics and Computing open to all presenters at MCMSki IV! We now need to send a reminder to them…


Filed under: Books, Mountains, pictures, R, Statistics, Travel, University life, Wines Tagged: BayesComp, Blossom skis, Chamonix-Mont-Blanc, ISBA, MCMSki, ski race, Statistics and Computing

Categories: Bayesian Bloggers

X’mas bookreads

Xian's Og - Fri, 2014-01-10 19:14

Even though I am beyond schedule at several levels of reality, I took some time off during the X’mas break to read a few of the books from my to-read pile. The first one was The Dirty Streets of Heaven by Tad Williams. While I read two fantasy series by Williams, Memory, Sorrow and Thorn, and Shadowmarch, which major drawback was that they both were unnecessarily long, this short novel is a mix of urban fantasy and of detective story, except that the detective working for Heaven in our current universe and fighting the “Opposition”, i.e. Hell, at every moment. This may sound quite a weird setting, but I nonetheless enjoyed the plot, the characters and the witty dialogues (as in “a man big enough to have his own zip code”). There were some lengthy parts, inevitably, but the whole scheme was addictive enough that I read it within two days. Now, there is a second (and then a third) volume in the series that does not sound up to par, judging from the amazon reviews. But this first volume got a very positive review from Patrick Rothfuss and it can be read on its own.

The second book I read over the vacations in Chamonix is Olen Steinhauer’s An American spy. This is the third instalment in the stories of Milo Weaver, the never-truly-retired Tourist. The volume is more into tying loose ends from previous books than into creating a new compelling story, even though it plays on the disappearance of loved ones and on a maze of double- and triple-agents. The fact that the story is told from many perspectives does not help (it is as if Weaver is now a secondary character) and the conclusion is fairly anticlimactic. A bit of nitpicking: a couple of spies (Tourists) travel to Jeddah in Saudi Arabia on a tourist visa, but there is no such thing as a Saudi tourist visa. Plus, the behaviour of the characters there is incompatible with the strict laws of Saudi Arabia.

A third book completed during those vacations is Gutted, by Tony Black. (I had actually bought this book in Warwick for my son’ British studies project but he did not look further than the backcover.) The book is taking place in Edinburgh, starting on Corstorphine Hill with a dog beating, and continuing in the seediest estates of Edinburgh where dog fights are parts of the shadow economy. The main character of the novel is the anti-hero Gus Drury, who is engaged so thoroughly in self-destruction that he would make John Rebus sound like a teetotaller! Gus is an ex-journalist who lost his job and wife to scoosh, running a pub with the help of two friends. Why he gets involved in an investigation remains unclear to me for the whole book: While Black has been hailed as a beacon for Celtic Noir, and while the style is gritty and enjoyable, I find the plot a wee bit shallow, with an uncomfortable number of coincidences. While finding this book was like discovering a long lost sibling of Rankin’s Rebus, with a pleasurable stroll through Edinburgh (!), I am far from certain I can contemplate reading the whole series

Lastly, I read (most of) Giant Thief, by David Tallerman. By bits. This may be the least convincing book in the list. The story is one of a thief who finds himself enrolled in an army he has no reason to support and steals an artefact which value he is unaware of when deserting, along with a giant. The pursuit drags on forever. There are many reasons I disliked the book: the plot is shallow, the main character is the ultimate cynic, with not enough depth to build upon. Definitely missing the sparkling charm of the Lies of Locke Lamorra.


Filed under: Books, Kids, Travel Tagged: Christmas, Edinburgh, Ian Rankin, Milo Weaver, Patrick Rothfuss, reading list, Scottish crime novel, Tad Williams, the lies of Locke Lamora, tourism
Categories: Bayesian Bloggers

MCMSki IV [day 3]

Xian's Og - Wed, 2014-01-08 19:14

Already on the final day..! And still this frustration in being unable to attend three sessions at once… Andrew Gelman started the day with a non-computational talk that broached on themes that are familiar to readers of his blog, on the misuse of significance tests and on recommendations for better practice. I then picked the Scaling and optimisation of MCMC algorithms session organised by Gareth Roberts, with optimal scaling talks by Tony Lelièvre, Alex Théry and Chris Sherlock, while Jochen Voss spoke about the convergence rate of ABC, a paper I already discussed on the blog. A fairly exciting session showing that MCMC’ory (name of a workshop I ran in Paris in the late 90′s!) is still well and alive!

After the break (sadly without the ski race!), the software round-table session was something I was looking for. The four softwares covered by this round-table were BUGS, JAGS, STAN, and BiiPS, each presented according to the same pattern. I would have like to see a “battle of the bands”, illustrating pros & cons for each language on a couple of models & datasets. STAN got the officious prize for cool tee-shirts (we should have asked the STAN team for poster prize tee-shirts). And I had to skip the final session for a flu-related doctor appointment…

I called for a BayesComp meeting at 7:30, hoping for current and future members to show up and discuss the format of the future MCMski meetings, maybe even proposing new locations on other “sides of the Italian Alps”! But (workshop fatigue syndrome?!), no-one showed up. So anyone interested in discussing this issue is welcome to contact me or David van Dyk, the new BayesComp program chair.


Filed under: Mountains, pictures, R, Statistics, Travel, University life Tagged: Alps, banquet, BayesComp, Bayesian model choice, BiiPS, BUGS, Chamonix-Mont-Blanc, flu, JAGS, MCMSki IV, ski race, software, STAN, statistical significance, tee-shirt
Categories: Bayesian Bloggers

MCMSki IV [day 2.5]

Xian's Og - Wed, 2014-01-08 09:14

Despite a good rest during the ski break, my cold did not get away (no magic left in this world!) and I thus had a low attention span to attend the Bayesian statistics and Population genetics session: while Jukka Corander mentioned the improvement brought by our AMIS algorithm, I had difficulties getting the nature of the model, if only because he used a blackboard-like font that made math symbols too tiny to read. (Nice fonts, otherwise!), Daniel Lawson (of vomiting Warhammer fame!) talked about the alluring notion of a statistical emulator, and Barbara Engelhardt talked about variable selection in a SNP setting. I did not get a feeling on how handling ten millions of SNPs was possible in towards a variable selection goal.  My final session of the day was actually “my” invited session on ABC methods, where Richard Everitt presented a way of mixing exact approximation with ABC and synthetic likelihood (Wood, Nature) approximations. The resulting MAVIS algorithm is  not out yet. The second speaker was Ollie Ratman, who spoke on his accurate ABC that I have discussed many times here. And Jean-Michel Marin managed to drive from Montpelier, just in time to deliver his talk on our various explorations of the ABC model choice problem.

After a quick raclette at “home”, we headed back to the second poster session, where I had enough of a clear mind and not too much of a headache (!) to have several interesting discussions, incl. a new parallelisation suggested  by Ben Calderhead, the sticky Metropolis algorithm of Luca Martino, the airport management video of Jegar Pitchforth, the mixture of Dirichlet distributions for extremes by Anne Sabourin, not mentioning posters from Warwick or Paris. At the end of the evening  I walked back to my apartment with the Blossom skis we had brought in the morning to attract registrations for the ski race: not enough to make up for the amount charged by the ski school. Too bad, especially given Anto’s efforts to get this amazing sponsoring!


Filed under: Mountains, pictures, Statistics, University life Tagged: ABC, AMIS, extremes, parallelisation, poster session, raclette, SNPs, sticky Metropolis, synthetic likelihood, warhammer
Categories: Bayesian Bloggers

MCMSki [day 2]

Xian's Og - Tue, 2014-01-07 19:14

I was still feeling poorly this morning with my brain in a kind of flu-induced haze so could not concentrate for a whole talk, which is a shame as I missed most of the contents of the astrostatistics session put together by David van Dyk… Especially the talk by Roberto Trotta I was definitely looking for. And the defence of nested sampling strategies for marginal likelihood approximations. Even though I spotted posterior distributions for WMAP and Plank data on the ΛCDM that reminded me of our own work in this area… Apologies thus to all speakers for dozing in and out, it was certainly not due to a lack of interest!

Sebastian Seehars mentioned emcee (for ensemble Monte Carlo), with a corresponding software nicknamed “the MCMC hammer”, and their own CosmoHammer software. I read the paper by Goodman and Ware (2010) this afternoon during the ski break (if not on a ski lift!). Actually, I do not understand why an MCMC should be affine invariant: a good adaptive MCMC sampler should anyway catch up the right scale of the target distribution. Other than that, the ensemble sampler reminds me very much of the pinball sampler we developed with Kerrie Mengersen (1995 Valencia meeting), where the target is the product of L targets,

and a Gibbs-like sampler can be constructed, moving one component (with index k, say) of the L-sample at a time. (Just as in the pinball sampler.) Rather than avoiding all other components (as in the pinball sampler), Goodman and Ware draw a single other component at random  (with index j, say) and make a proposal away from it:

where ζ is a scale random variable with (log-) symmetry around 1. The authors claim improvement over a single track Metropolis algorithm, but it of course depends on the type of Metropolis algorithms that is chosen… Overall, I think the criticism of the pinball sampler also applies here: using a product of targets can only slow down the convergence. Further, the affine structure of the target support is not a given. Highly constrained settings should not cope well with linear transforms and non-linear reparameterisations would be more efficient….


Filed under: Mountains, pictures, Statistics, University life Tagged: Alps, astrostatistics, Bayesian model choice, Chamonix-Mont-Blanc, cosmology, emcee, flu, ΛCDM model, nested sampling, winter sports
Categories: Bayesian Bloggers

MCMSki IV [day 1.5]

Xian's Og - Tue, 2014-01-07 09:14

The afternoon sessions I attended were “Computational and Methodological Challenges in evidence synthesis and multi-step” organised by Nicky Best and Sylvia Richardson and “Approximate inference” put together by Dan Simpson. Since both Nicky and Sylvia were alas unable to attend MCMSki, I chaired their session, which I found most interesting as connected to a recurrent questioning of mine about conducting inference with partial likelihoods. Chris Paciorek also introduced a new software, Nimble, that he is currently developing.  There will be a round table on Wednesday on MCMC related software, after the ski race, so will wait till then before commenting on my reticence to engage into new softwares like Stan or Nimble… Dan’s session was also closely related with my interests, esp. Nicolas Chopin talking about quasi-Monte-Carlo versions of SMC and Clare McGrory mixing variational Bayes and sequential Monte-Carlo. In the setting of mixtures, Clare mentioned using variational Bayes as a way of estimating the number of components, which somewhat surprises me in a sequential framework because it is likely to underestimate the true number…

The poster session was successful, I think, even though a growing fever (from a cold caught on a freezing lift on Sunday!) prevented me from appreciating it fully. As usual, I wish I had had more time to discuss with all of the 35 poster presenters. (I still enjoyed very much discussing ABC at length with Jukka Corander and his students. And seeing some of my PhD students giving their first poster ever. While managing to cook a decent scallop risotto in the dinner break.) Any contribution of  a guest post on the poster sessions is most welcome!

Overall, a very full day (which ended up to late to include this part in the summary of my sessions). With very few cancellations due to the bad weather. And a very enjoyable setting in the Majestic (former) palace! Incidentally, it seems like the periphrase “French side of the Italian Alps” Antonietta Mira used for the first announcement of the conference is getting into a recurrent joke, with new variants (“Swiss side of the Italian Alps”, &tc.)


Filed under: Mountains, pictures, R, Statistics, Travel, University life Tagged: ABC, Chamonix-Mont-Blanc, fever, MCMSki IV, Nimble, poster session, Richard Tweedie
Categories: Bayesian Bloggers

MCMSki IV [day 1]

Xian's Og - Mon, 2014-01-06 19:14

The first day of MCMSki IV went by rather smoothly, I do believe, as most speakers were there, with just a dozen participants missing. And no-one broke a limb or vanished over a cliff when skiing. The only trouble I had was to pick between the parallel sessions, a definitive drawback of the format I had supported from the start… Any proposal for a guest post from participants is welcomed!!!

Chris Holmes gave an exciting first plenary talk, in the spirit of the “substitute” prior talk Stephen Walker had given in London for Bayes 250. This time, Chris talked about robustifying Bayesian inference by creating Kullback-Leibler neighbourhoods and taking least favourable priors within these neighbourhoods, least favourable in a decision-theoretic sense based on a loss function. While this approach has the drawback of being based upon a minimax principle, and requires the determination of a loss function, I find it nonetheless very appealing. And recovering tempered distributions is just cool! (The paper should be arXived within days.)

I then went to the Convergence of MCMC algorithms session, with Gallin Jones and Jim Hobert presenting uniform ergodicity results. Jim analysed the approach to the logistic regression model advocated by Nick Polson, James Scott, and Jesse Windle, which brings a parallel to the older latent variable solution of Albert and Chib (1993). Krys Łatuszyński showed that the spectral gap does not vary between the deterministic, random and random order Gibbs samplers (with a maybe tongue-in-cheek title Solidarity in the title that could refer to the iconic Solidarność… Or not.) And Éric Moulines established the uniformly ergodicity of a particle filter algorithm, showing that the number of particles had to grow as T1+ε for any positive ε. (And managed to get the paper arXived today!)


Filed under: Mountains, pictures, Statistics, Travel, University life
Categories: Bayesian Bloggers

Blossom skis [Tweedie race prize]

Xian's Og - Mon, 2014-01-06 09:14

  

Here are the two pairs of beautiful skies offered by Blossom for the Richard Tweedie ski race of Wednesday! Provided the snow cover holds till then on the ski track. So far, the chances are very good, according to the ski school organisers. Confirmation this afternoon! It will definitely take part: registration tomorrow morning (Jan. 7, closing at half past noon) and meeting for the race at the top of the Parsa ski-lift (reached via the Brévent cable-car) on the stade (stadium) from 1pm onwards.


Filed under: Mountains, Running, University life Tagged: Blossom skis, Chamonix-Mont-Blanc, ESF, MCMSki IV, Richard Tweedie
Categories: Bayesian Bloggers