“The collection of estimates may be determined by questions such as: How much storage is available? Can all the data be kept in memory or only a subset? How much processing power is available? Are there parallel or distributed systems that can be exploited?”
Daniel Sussman, Alexander Volfovsky, and Edoardo Airoldi from Harvard wrote a very interesting paper about setting a balance between statistical efficiency and computational efficiency, a theme that resonates with our recent work on ABC and older considerations about the efficiency of Monte Carlo algorithms. While the paper avoids drifting towards computer science even with a notion like algorithmic complexity, I like the introduction of a loss function in the comparison game, even though the way to combine both dimensions is unclear. And may limit the exercise to an intellectual game. In an ideal setting one would set the computational time, like “I have one hour to get this estimate”, and compare risks under that that computing constraint. Possibly dumping some observations from the sample to satisfy the constraint. Ideally. Which is why this also reminds me of ABC: given an intractable likelihood, one starts by throwing away some data precision by using a tolerance ε and usually more through an insufficient statistic. Hence ABC procedures could also be compared in such terms.
In the current paper, the authors only compare schemes of breaking the sample into bits to handle each observation only once. Meaning it cannot be used in both the empirical mean and the empirical variance. This sounds a bit contrived in that the optimum allocation depends on the value of the parameter the procedure attempts to estimate. Still, it could lead to a new form of bandit problems: given a bandit with as many arms as there are parameters, at each new observation, decide on the allocation towards minimising the overall risk. (There is a missing sentence at the end of Section 4.)
Any direction for turning those considerations into a practical decision machine would be fantastic, although the difficulties are formidable, from deciding between estimators and selecting a class of estimators, to computing costs and risks depending on unknown parameters.
Filed under: Books, pictures, Statistics, University life Tagged: ABC, computing cost, efficiency, Harvard University, insufficiency, limited resources, multi-armed bandits
[Here is a call for a special issue on Bayesian nonparametrics, edited by Alessio Benavoli , Antonio Lijoi and Antonietta Mira, for an Elsevier journal I had never heard of previously:]
The International Journal of Approximate Reasoning is pleased to announce a special issue on “Bayesian Nonparametrics”. The submission deadline is *December 1st*, 2015.
The aim of this Special Issue is twofold. First, it is to give a broad overview of the most popular models used in BNP and their application in
Artificial Intelligence, by means of tutorial papers. Second, the Special Issue will focus on theoretical advances and challenging applications of BNP with special emphasis on the following aspects:
- Methodological and theoretical developments of BNP
- Treatment of imprecision and uncertainty with/in BNP methods
- Formal applications of BNP methods to novel applied problems
- New computational and simulation tools for BNP inference.
Filed under: Books, Statistics, University life Tagged: Bayesian non-parametrics, Elsevier, International Journal of Approximate Reasoning
Another arXived paper in the recent series about big or tall data and how to deal with it by MCMC. Which pertains to the embarrassingly parallel category. As in the previously discussed paper, the authors (Xiangyu Wang, Fangjian Guo, Katherine Heller, and David Dunson) chose to break the prior itself into m bits… (An additional point from last week criticism is that, were an unbiased estimator of each term in the product available in an independent manner, the product of the estimators would be the estimator of the product.) In this approach, the kernel estimator of Neiswanger et al. is replaced with a random partition tree histogram. Which uses the same block partition across all terms in the product representation of the posterior. And hence ends up with a smaller number of terms in the approximation, since it does not explode with m. (They could have used Mondrian forests as well! However I think their quantification of the regular kernel method cost as an O(Tm) approach does not account for Neiswanger et al.’s trick in exploiting the product of kernels…) The so-called tree estimate can be turned into a random forest by repeating the procedure several times and averaging. The simulation comparison runs in favour of the current method when compared with other consensus or non-parametric methods. Except in the final graph (Figure 5) which shows several methods achieving the same prediction accuracy against running time.
Filed under: Books, pictures, Statistics, University life Tagged: big data, embarassingly parallel, huge data, MCMC, Mondrian forests, parallel MCMC, random partition trees, tall data
Dalhin, Villani [Mattias, not Cédric] and Schön arXived a paper this week with the above title. The type of intractable likelihood they consider is a non-linear state-space (HMM) model and the SMC-ABC they propose is based on an optimised Laplace approximation. That is, replacing the posterior distribution on the parameter θ with a normal distribution obtained by a Taylor expansion of the log-likelihood. There is no obvious solution for deriving this approximation in the case of intractable likelihood functions and the authors make use of a Bayesian optimisation technique called Gaussian process optimisation (GPO). Meaning that the Laplace approximation is the Laplace approximation of a surrogate log-posterior. GPO is a Bayesian numerical method in the spirit of the probabilistic numerics discussed on the ‘Og a few weeks ago. In the current setting, this means iterating three steps
- derive an approximation of the log-posterior ξ at the current θ using SMC-ABC
- construct a surrogate log-posterior by a Gaussian process using the past (ξ,θ)’s
- determine the next value of θ
In the first step, a standard particle filter cannot be used to approximate the observed log-posterior at θ because the conditional density of observed given latent is intractable. The solution is to use ABC for the HMM model, in the spirit of many papers by Ajay Jasra and co-authors. However, I find the construction of the substitute model allowing for a particle filter very obscure… (A side effect of the heat wave?!) I can spot a noisy ABC feature in equation (7), but am at a loss as to how the reparameterisation by the transform τ is compatible with the observed-given-latent conditional being unavailable: if the pair (x,v) at time t has a closed form expression, so does (x,y), at least on principle, since y is a deterministic transform of (x,v). Another thing I do not catch is why having a particle filter available prevent the use of a pMCMC approximation.
The second step constructs a Gaussian process posterior on the log-likelihood, with Gaussian errors on the ξ’s. The Gaussian process mean is chosen as zero, while the covariance function is a Matérn function. With hyperparameters that are estimated by maximum likelihood estimators (based on the argument that the marginal likelihood is available in closed form). Turning the approach into an empirical Bayes version.
The next design point in the sequence of θ’s is the argument of the maximum of a certain acquisition function, which is chosen here as a sort of maximum regret associated with the posterior predictive associated with the Gaussian process. With possible jittering. At this stage, it reminded me of the Gaussian process approach proposed by Michael Gutmann in his NIPS poster last year.
Overall, the method is just too convoluted for me to assess its worth and efficiency without a practical implementation to… practice upon, for which I do not have time! Hence I would welcome any comment from readers having attempted such implementations. I also wonder at the lack of link with Simon Wood‘s Gaussian approximation that appeared in Nature (2010) and was well-discussed in the Read Paper of Fearnhead and Prangle (2012).
Filed under: Books, pictures, Statistics, University life Tagged: ABC, Gaussian processes, Laplace approximation, Linköping University, Matérn covariance function, maximum likelihood estimation, noisy ABC, pMCMC, SMC-ABC, state space model, stochastic volatility, Sweden, Uppsala University
Filed under: Books, pictures, Travel, University life, Wines Tagged: France, La Comédie, Montpellier, sunset, theatre
- Bayesian Survival Model Based on Moment Characterization by Arbel, Julyan et al.
- A New Finite Approximation for the NGG Mixture Model: An Application to Density Estimation by Bianchini, Ilaria
- Distributed Estimation of Mixture Model by Dedecius, Kamil et al.
- Jeffreys’ Priors for Mixture Estimation by Grazian, Clara and X
- A Subordinated Stochastic Process Model by Palacios, Ana Paula et al.
- Bayesian Variable Selection for Generalized Linear Models Using the Power-Conditional-Expected-Posterior Prior by Perrakis, Konstantinos et al.
- Application of Interweaving in DLMs to an Exchange and Specialization Experiment by Simpson, Matthew
- On Bayesian Based Adaptive Confidence Sets for Linear Functionals by Szabó, Botond
- Identifying the Infectious Period Distribution for Stochastic Epidemic Models Using the Posterior Predictive Check by Alharthi, Muteb et al.
- A New Strategy for Testing Cosmology with Simulations by Killedar, Madhura et al.
- Formal and Heuristic Model Averaging Methods for Predicting the US Unemployment Rate by Kolly, Jeremy
- Bayesian Estimation of the Aortic Stiffness based on Non-invasive Computed Tomography Images by Lanzarone, Ettore et al.
- Bayesian Filtering for Thermal Conductivity Estimation Given Temperature Observations by Martín-Fernández, Laura et al.
- A Mixture Model for Filtering Firms’ Profit Rates by Scharfenaker, Ellis et al.
Filed under: Books, Kids, pictures, Statistics, Travel, University life, Wines Tagged: Austria, BAYSM 2014, conference, proceedings, Springer-Verlag, Vienna, Wien, WU Wirtschaftsuniversität Wien, young Bayesians
While I now try to abstain from participating to the Cross Validated forum, as it proves too much of a time-consuming activity with little added value (in the sense that answers are much too often treated as disposable napkins by users who cannot be bothered to open a textbook and who usually do not exhibit any long-term impact of the provided answer, while clogging the forum with so many questions that the individual entries seem to get so little traffic, when compared say with the stackoverflow forum, to the point of making the analogy with disposable wipes more appropriate!), I came across a truly interesting question the other night. Truly interesting for me in that I had never considered the issue before.
The question is essentially wondering at how to simulate from a distribution defined by its failure rate function, which is connected with the density f of the distribution by
From a purely probabilistic perspective, defining the distribution through f or through η is equivalent, as shown by the relation
but, from a simulation point of view, it may provide a different entry. Indeed, all that is needed is the ability to solve (in X) the equation
when U is a Uniform (0,1) variable. Which may help in that it does not require a derivation of f. Obviously, this also begs the question as to why would a distribution be defined by its failure rate function.
Filed under: Books, Kids, Statistics, University life Tagged: cross validated, failure rate, Monte Carlo Statistical Methods, probability theory, reliability, simulation, StackExchange, stackoverflow, survival analysis
[A rather stinky piece in The Guardian today, written by a consultant self-styled Higher Education expert… No further comments needed!]
“The reasons cited for this laggardly response [to innovations] will be familiar to any observer of the university system: an inherently conservative and risk-averse culture in most institutions; sclerotic systems and processes designed for a different world, and a lack of capacity, skills and willingness to change among an ageing academic community. All these are reinforced by perceptions that most proposed innovations are over-hyped and that current ways of operating have plenty of life left in them yet.”
Filed under: Books, Kids, pictures, University life Tagged: marketing, privatisation, reform, The Guardian, United Kingdom
Heiko Strathmann, Dino Sejdinovic, Samuel Livingstone, Zoltán Szabó, and Arthur Gretton arXived paper about Kamiltonian MCMC generated comments from Michael Betancourt, Dan Simpson and myself, which themselves induced the following reply by Heiko, detailed enough to deserve a post of its own.
Adaptation and ergodicity.
We certainly agree that the naive approach of using a non-parametric kernel density estimator on the chain history (as in [Christian’s book, Example 8.8]) as a *proposal* fails spectacularly on simple examples: the probability of proposing in unexplored regions is extremely small, independent of the current position of the MCMC trajectory. This is not what we do though. Instead, we use the gradient of a density estimator, and not the density itself, for our HMC proposal. Just like KAMH, KMC lite in fact falls back to Random Walk Metropolis in previously unexplored regions and therefore inherits geometric ergodicity properties. This in particular includes the ability to explore previously “unseen” regions, even if adaptation has stopped. I implemented a simple illustration and comparison here.
The main point of the ABC example, is that our method does not suffer from the additional bias from Gaussian synthetic likelihoods when being confronted with skewed models. But there is also a computational efficiency aspect. The scheme by Meeds et al. relies on finite differences and requires $2D$ simulations from the likelihood *every time* the gradient is evaluated (i.e. every leapfrog iteration) and H-ABC discards this valuable information subsequently. In contrast, KMC accumulates gradient information from simulations: it only requires to simulate from the likelihood *once* in the accept/reject step after the leapfrog integration (where gradients are available in closed form). The density is only updated then, and not during the leapfrog integration. Similar work on speeding up HMC via energy surrogates can be applied in the tall data scenario.
Monte Carlo gradients.
Approximating HMC when gradients aren’t available is in general a difficult problem. One approach (like surrogate models) may work well in some scenarios while a different approach (i.e. Monte Carlo) may work better in others, and the ABC example showcases such a case. We very much doubt that one size will fit all — but rather claim that it is of interest to find and document these scenarios.
Michael raised the concern that intractable gradients in the Pseudo-Marginal case can be avoided by running an MCMC chain on the joint space (e.g. $(f,\theta)$ for the GP classifier). To us, however, the situation is not that clear. In many cases, the correlations between variables can cause convergence problems (see e.g. here) for the MCMC and have to be addressed by de-correlation schemes (as here), or e.g. by incorporating geometric information, which also needs fixes as Michaels’s very own one. Which is the method of choice with a particular statistical problem at hand? Which method gives the smallest estimation error (if that is the goal?) for a given problem? Estimation error per time? A thorough comparison of these different classes of algorithms in terms of performance related to problem class would help here. Most papers (including ours) only show experiments favouring their own method.
GP estimator quality.
Finally, to address Michael’s point on the consistency of the GP estimator of the density gradient: this is discussed In the original paper on the infinite dimensional exponential family. As Michael points out, higher dimensional problems are unavoidably harder, however the specific details are rather involved. First, in terms of theory: both the well-specified case (when the natural parameter is in the RKHS, Section 4), and the ill-specified case (the natural parameter is in a “reasonable”, larger class of functions, Section 5), the estimate is consistent. Consistency is obtained in various metrics, including the L² error on gradients. The rates depend on how smooth the natural parameter is (and indeed a poor choice of hyper-parameter will mean slower convergence). The key point, in regards to Michael’s question, is that the smoothness requirement becomes more restrictive as the dimension increases: see Section 4.2, “range space assumption”.
Second, in terms of practice: we have found in experiments that the infinite dimensional exponential family does perform considerably better than a kernel density estimator when the dimension increases (Section 6). In other words, our density estimator can take advantage of smoothness properties of the “true” target density to get good convergence rates. As a practical strategy for hyper-parameter choice, we cross-validate, which works well empirically despite being distasteful to Bayesians. Experiments in the KMC paper also indicate that we can scale these estimators up to dimensions in the 100s on Laptop computers (unlike most other gradient estimation techniques in HMC, e.g. the ones in your HMC & sub-sampling note, or the finite differences in Meeds et al).
Filed under: Books, Statistics, University life Tagged: adaptive MCMC methods, Bayesian quadrature, Gatsby, Hamiltonian Monte Carlo, London, Markov chain, Monte Carlo Statistical Methods, non-parametric kernel estimation, reproducing kernel Hilbert space, RKHS, smoothness
“Unfortunately, the factorization does not make it immediately clear how to aggregate on the level of samples without first having to obtain an estimate of the densities themselves.” (p.2)
The recently arXived variational consensus Monte Carlo is a paper by Maxim Rabinovich, Elaine Angelino, and Michael Jordan that approaches the consensus Monte Carlo principle from a variational perspective. As in the embarrassingly parallel version, the target is split into a product of K terms, each being interpreted as an unnormalised density and being fed to a different parallel processor. The most natural partition is to break the data into K subsamples and to raise the prior to the power 1/K in each term. While this decomposition makes sense from a storage perspective, since each bit corresponds to a different subsample of the data, it raises the question of the statistical pertinence of splitting the prior and my feelings about it are now more lukewarm than when I commented on the embarrassingly parallel version, mainly for the reason that it is not reparameterisation invariant—getting different targets if one does the reparameterisation before or after the partition—and hence does not treat the prior as the reference measure it should be. I therefore prefer the version where the same original prior is attached to each part of the partitioned likelihood (and even more the random subsampling approaches discussed in the recent paper of Bardenet, Doucet, and Holmes). Another difficulty with the decomposition is that a product of densities is not a density in most cases (it may even be of infinite mass) and does not offer a natural path to the analysis of samples generated from each term in the product. Nor an explanation as to why those samples should be relevant to construct a sample for the original target.
“The performance of our algorithm depends critically on the choice of aggregation function family.” (p.5)
Since the variational Bayes approach is a common answer to complex products models, Rabinovich et al. explore the use of variational Bayes techniques to build the consensus distribution out of the separate samples. As in Scott et al., and Neiswanger et al., the simulation from the consensus distribution is a transform of simulations from each of the terms in the product, e.g., a weighted average. Which determines the consensus distribution as a member of an aggregation family defined loosely by a Dirac mass. When the transform is a sum of individual terms, variational Bayes solutions get much easier to find and the authors work under this restriction… In the empirical evaluation of this variational Bayes approach as opposed to the uniform and Gaussian averaging options in Scott et al., it improves upon those, except in a mixture example with a large enough common variance.
In fine, despite the relevance of variational Bayes to improve the consensus approximation, I still remain unconvinced about the use of the product of (pseudo-)densities and the subsequent mix of simulations from those components, for the reason mentioned above and also because the tail behaviour of those components is not related with the tail behaviour of the target. Still, this is a working solution to a real problem and as such is a reference for future works.
Filed under: Books, Statistics, University life Tagged: big data, consensus Monte Carlo, embarassingly parallel, large data problems, subsampling, tall data, variational Bayes methods
Following my earlier comments on Alexander Ly, Josine Verhagen, and Eric-Jan Wagenmakers, from Amsterdam, Joris Mulder, a special issue editor of the Journal of Mathematical Psychology, kindly asked me for a written discussion of that paper, discussion that I wrote last week and arXived this weekend. Besides the above comments on ToP, this discussion contains some of my usual arguments against the use of the Bayes factor as well as a short introduction to our recent proposal via mixtures. Short introduction as I had to restrain myself from reproducing the arguments in the original paper, for fear it would jeopardize its chances of getting published and, who knows?, discussed.
Filed under: Books, Kids, pictures, Running, Statistics, Travel, University life Tagged: Amsterdam, Bayes factor, boat, Harold Jeffreys, Holland, Journal of Mathematical Psychology, psychometrics, sunrise, Theory of Probability, XXX
While cooking for a late Sunday lunch today [sweet-potatoes röstis], I was listening as usual to the French Public Radio (France Inter) and at some point heard the short [10mn] Périphéries that gives every weekend an insight on the suburbs [on the “other side’ of the Parisian Périphérique boulevard]. The idea proposed by a geographer from Montpellier, Emmanuel Vigneron, was to point out the health inequalities between the wealthy 5th arrondissement of Paris and the not-so-far-away suburbs, by following the RER B train line from Luxembourg to La Plaine-Stade de France…
The disparities between the heart of Paris and some suburbs are numerous and massive, actually the more one gets away from the lifeline represented by the RER A and RER B train lines, so far from me the idea of negating this opposition, but the presentation made during those 10 minutes of Périphéries was quite approximative in statistical terms. For instance, the mortality rate in La Plaine is 30% higher than the mortality rate in Luxembourg and this was translated into the chances for a given individual from La Plaine to die in the coming year are 30% higher than if he [or she] lives in Luxembourg. Then a few minutes later the chances for a given individual from Luxembourg to die are 30% lower than he [or she] lives in La Plaine…. Reading from the above map, it appears that the reference is the mortality rate for the Greater Paris. (Those are 2010 figures.) This opposition that Vigneron attributes to a different access to health facilities, like the number of medical general practitioners per inhabitant, does not account for the huge socio-demographic differences between both places, for instance the much younger and maybe larger population in suburbs like La Plaine. And for other confounding factors: see, e.g., the equally large difference between the neighbouring stations of Luxembourg and Saint-Michel. There is no socio-demographic difference and the accessibility of health services is about the same. Or the similar opposition between the southern suburban stops of Bagneux and [my local] Bourg-la-Reine, with the same access to health services… Or yet again the massive decrease in the Yvette valley near Orsay. The analysis is thus statistically poor and somewhat ideologically biased in that I am unsure the data discussed during this radio show tells us much more than the sad fact that suburbs with less favoured populations show a higher mortality rate.
Filed under: Statistics, Travel Tagged: Bagneux, boulevard périphérique, Bourg-la-Rein, France, France Inter, inequalities, Luxembourg, national public radio, Orsay, Paris, Paris suburbs, Périphéries, RER B, Saint-Michel, Stade de France, Yvette
Heiko Strathmann, Dino Sejdinovic, Samuel Livingstone, Zoltán Szabó, and Arthur Gretton arXived a paper last week about Kamiltonian MCMC, the K being related with RKHS. (RKHS as in another KAMH paper for adaptive Metropolis-Hastings by essentially the same authors, plus Maria Lomeli and Christophe Andrieu. And another paper by some of the authors on density estimation via infinite exponential family models.) The goal here is to bypass the computation of the derivatives in the moves of the Hamiltonian MCMC algorithm by using a kernel surrogate. While the genuine RKHS approach operates within an infinite exponential family model, two versions are proposed, KMC lite with an increasing sequence of RKHS subspaces, and KMC finite, with a finite dimensional space. In practice, this means using a leapfrog integrator with a different potential function, hence with a different dynamics.
The estimation of the infinite exponential family model is somewhat of an issue, as it is estimated from the past history of the Markov chain, simplified into a random subsample from this history [presumably without replacement, meaning the Markovian structure is lost on the subsample]. This is puzzling because there is dependence on the whole past, which cancels ergodicity guarantees… For instance, we gave an illustration in Introducing Monte Carlo Methods with R [Chapter 8] of the poor impact of approximating the target by non-parametric kernel estimates. I would thus lean towards the requirement of a secondary Markov chain to build this kernel estimate. The authors are obviously aware of this difficulty and advocate an attenuation scheme. There is also the issue of the cost of a kernel estimate, in O(n³) for a subsample of size n. If, instead, a fixed dimension m for the RKHS is selected, the cost is in O(tm²+m³), with the advantage of a feasible on-line update, making it an O(m³) cost in fine. But again the worry of using the whole past of the Markov chain to set its future path…
Among the experiments, a KMC for ABC that follows the recent proposal of Hamiltonian ABC by Meeds et al. The arguments are interesting albeit sketchy: KMC-ABC does not require simulations at each leapfrog step, is it because the kernel approximation does not get updated at each step? Puzzling.
I also discussed the paper with Michael Betancourt (Warwick) and here his comments:
“I’m hesitant for the same reason I’ve been hesitant about algorithms like Bayesian quadrature and GP emulators in general. Outside of a few dimensions I’m not convinced that GP priors have enough regularization to really specify the interpolation between the available samples, so any algorithm that uses a single interpolation will be fundamentally limited (as I believe is born out in non-trivial scaling examples) and trying to marginalize over interpolations will be too awkward.
They’re really using kernel methods to model the target density which then gives the gradient analytically. RKHS/kernel methods/ Gaussian processes are all the same math — they’re putting prior measures over functions. My hesitancy is that these measures are at once more diffuse than people think (there are lots of functions satisfying a given smoothness criterion) and more rigid than people think (perturb any of the smoothness hyper-parameters and you get an entirely new space of functions).
When using these methods as an emulator you have to set the values of the hyper-parameters which locks in a very singular definition of smoothness and neglects all others. But even within this singular definition there are a huge number of possible functions. So when you only have a few points to constrain the emulation surface, how accurate can you expect the emulator to be between the points?
In most cases where the gradient is unavailable it’s either because (a) people are using decades-old Fortran black boxes that no one understands, in which case there are bigger problems than trying to improve statistical methods or (b) there’s a marginalization, in which case the gradients are given by integrals which can be approximated with more MCMC. Lots of options.”
Filed under: Books, Statistics, University life Tagged: adaptive MCMC methods, Bayesian quadrature, Gatsby, Hamiltonian Monte Carlo, Introducing Monte Carlo Methods with R, London, Markov chain, non-parametric kernel estimation, reproducing kernel Hilbert space, RKHS, smoothness
Filed under: pictures, Travel, University life Tagged: bike, tangerine, University of Warwick, winter
When visiting a bookstore in Florence last month, during our short trip to Tuscany, I came upon this book with enough of a funny cover and enough of a funny title (possibly capitalising on the similarity with “the girl who played with fire”] to make me buy it. I am glad I gave in to this impulse as the book is simply hilarious! The style and narrative relate rather strongly to the series of similarly [mostly] hilarious picaresque tales written by Paasilina and not only because both authors are from Scandinavia. There is the same absurd feeling that the book characters should not have this sort of things happening to them and still the morbid fascination to watch catastrophe after catastrophe being piled upon them. While the story is deeply embedded within the recent history of South Africa and [not so much] of Sweden for the past 30 years, including major political figures, there is no true attempt at making the story in the least realistic, which is another characteristic of the best stories of Paasilina. Here, a young girl escapes the poverty of the slums of Soweto, to eventually make her way to Sweden along with a spare nuclear bomb and a fistful of diamonds. Which alas are not eternal… Her intelligence helps her to overcome most difficulties, but even her needs from time to time to face absurd situations as another victim. All is well that ends well for most characters in the story, some of whom one would prefer to vanish in a gruesome accident. Which seemed to happen until another thread in the story saved the idiot. The satire of South Africa and of Sweden is most enjoyable if somewhat easy! Now I have to read the previous volume in the series, The Hundred-Year-Old Man Who Climbed Out of the Window and Disappeared!
Filed under: Books, Kids, Travel Tagged: Arto Paasilinna, book review, Finland, Firenze, Italy, Scandinavia, South Africa, Soweto, Sweden, the girl who saved the king of Sweden, The Hundred-Year-Old Man Who Climbed Out of the Window and Disappeared, Tuscany
Here are the download figures for my e-book with George as sent to me last week by my publisher Springer-Verlag. With an interesting surge in the past year. Maybe simply due to new selling strategies of the published rather to a wider interest in the book. (My royalties have certainly not increased!) Anyway thanks to all readers. As an aside for wordpress wannabe bloggers, I realised it is now almost impossible to write tables with WordPress, another illustration of the move towards small-device-supported blogs. Along with a new annoying “simpler” (or more accurately dumber) interface and a default font far too small for my eyesight. So I advise alternatives to wordpress that are more sympathetic to maths contents (e.g., using MathJax) and comfortable editing.
And the same for the e-book with Jean-Michel, which only appeared in late 2013. And contains more chapters than Introduction to Monte Carlo methods with R. Incidentally, a reader recently pointed out to me the availability of a pirated version of The Bayesian Choice on a Saudi (religious) university website. And of a pirated version of Introducing Monte Carlo with R on a Saõ Paulo (Brazil) university website. This may be alas inevitable, given the diffusion by publishers of e-chapters that can be copied with no limitations…
Filed under: Books, R, Statistics, University life Tagged: Bayesian Essentials with R, book sales, Brazil, copyright, Introduction to Monte Carlo Methods with R, Saudi Arabia, Springer-Verlag
As I was putting together a proposal for a special ABC session at ISBA 2016 in Santa Margherita di Pula, Sardinia, I received worried replies from would-be participants about the affordability of the meeting given their research funds! Since I had a similar worry with supporting myself and several of my PhD students, I looked around for low-cost alternatives and [already] booked a nearby villa in Santa Margherita di Pula for about 100€ per person for the whole week. Including bikes. Plus, several low-cost airlines like Easy Jet and Ryanair fly to Cagliari from European cities like Berlin, Paris, London, Geneva, and most Italian cities, for less than 100€ round-trip [with enough advanced planning] and if one really is on half a shoestring, there are regular buses connecting Cagliari to Santa Margherita di Pula for a few euros. This means in the end that supporting a PhD student or a postdoc within 5 years of a Ph.D. to attend ISBA 2016 from Europe can be budgeted at as low as a tight 500€ under limited funding resources, including the registration fees of 290€… So definitely affordable with long-term planning!
Filed under: Kids, pictures, Travel, University life, Wines Tagged: ABC, Bayesian conference, budget, Cagliari, pink flamingos, registration fees, research grant, Santa Margherita di Pula, Sardinia
Following yesterday’s surprise at the unpleasant conference business run by WASET, I was once again confronted today with conference fees that sound like an unacceptable siphoning of research funds and public money. One of my PhD students got earlier personally invited to present a talk at EUSIPCO 2015, a European signal processing conference taking place in Nice next September and she accepted the invitation. Now, contrary to yesterday’s example, this EUSIPCO 2015 is a genuine conference sponsored by several European signal processing societies. From what I understand, speakers and poster presenters must submit papers that are reviewed and then published in the conference proceedings, part of the IEEE Xplore on-line digital library (impact factor of 0.04). As the conference is drawing near, my student is asked to register and is “reminded” of small prints in the conference rules, namely that “at least one author per paper must register by June 19, 2015 at the full rate”, student or not student, which means a 300€ difference in the fees and has absolutely no justification whatsoever since the papers are only processed electronically…
I checked across a few of the past editions of EUSIPCO and the same rip-off rule applies to those as well. I see no rational explanation for this rule that sounds like highway robbery and leads to the de facto exclusion of students from conferences… In fine, my student withdrew her paper and participation at EUSIPCO.
Filed under: Kids, University life Tagged: conference fees, EUSIPCO 2015, IEEE, IEEE Xplore, Nice, signal processing, student fees