Archive for London

Dan Leno & the Limehouse Golem [book review]

Posted in Books, Travel with tags , , , , , , , , on November 26, 2017 by xi'an

Another book that came to my bedside rather randomly! It is in fact a 1994 book by Peter Ackroyd, not to be confused with Roger Ackroyd, a mystery book by Agatha Christie I remember reading in my teenage years! And takes place in Victorian London, around a woman Elisabeth Cree, who is a music hall celebrity and stands accused of murdering her husband. With the background of a series of gratuitous and inexplicable murders soon attributed to a supernatural creature. Called a golem for its ability to appear and vanish with no witness… There is a great idea in the plot but its implementation is quite tedious, with a plodding style that makes the conclusion a very long wait. This is not helped by Ackroyd borrowing so much from the life of a few well-known historical characters like Karl Marx, George Gissing, Dan Leno and Charles Babbage himself! Simply because they truly existed does not make these characters particularly exciting within the plot. Especially Babbage and his difference engine. (Which was exploited in a much better steampunk novel by William Gibson and Bruce Sterling!) The worst part is when Ackroyd reflects in the book on the engine being a “forerunner of the modern computer”, ruining the whole perspective. As I do not want to get into spoilers about the almost unexpected twists in the conclusion, let me conclude with quotes attributed to Babbage (or followers) about social statistics, for which he had devised the analytical engine.

“To be exactly informed about the lot of humankind (…) is to create the conditions in which it can be ameliorated. We must know before we can understand, and statistic evidence is the surest form of evidence currently in our possession.” (p.113)

“…the errors which arise from unsound reasoning neglecting true data are far more numerous and more durable than those which result from the absence of facts.” (p.119)

 

more positions in the UK [postdoc & professor]

Posted in Statistics with tags , , , , , , , , , , , on October 13, 2017 by xi'an

I have received additional emails from England advertising for positions in Bristol, Durham, and London, so here they are, with links to the complete advertising!

  1. The University of Bristol is seeking to appoint a number of Chairs in any areas of Mathematics or Statistical Science, in support of a major strategic expansion of the School of Mathematics. Deadline is December 4.
  2. Durham University is opening a newly created position of Professor of Statistics, with research and teaching duties. Deadline is November 6.
  3. Oliver Ratman, in the Department of Mathematics at Imperial College London, is seeking a Research Associate in Statistics and Pathogen Phylodynamics. Deadline is October 30.

wanton and furious cycling

Posted in pictures, Running, Travel with tags , , , , , , , , on September 10, 2017 by xi'an

A cyclist was convicted of “wanton or furious driving” last week in London after hitting a pedestrian crossing the street, leading to her death a few days later. The main legal argument for the conviction was that the cyclist was riding a “fixie”,  a bike with no front brake and fixed-gear, as used in track cycling. Which is illegal in Britain and, I just found out, in France too. (He was actually facing manslaughter, for which he got acquitted.) This is a most tragic accident, alas leading to a loss of a human life, and I did not look at the specifics, but I do not get the argument about the brakes and the furious driving: if the rider was going at about 28 km/h, which seems a reasonable speed in low density areas [and is just above my average speed in suburban Paris], and if the pedestrian stepped in his path six meters ahead, he had less than a second to react. Front brake or not, I am certainly unable to react and stop in this interval. And braking hard with the front brake will invariably lead to going over the bars: happens to me every time I have to stop for a car with my road bike. And would if I had to stop for a pedestrian.

Incidentally [or accidentally], here is the item of British Law from 1861 on which prosecution was based:

“Whosoever, having the charge of any carriage or vehicle, shall by wanton or furious driving or racing, or other wilful misconduct, or by wilful neglect, do or cause to be done any bodily harm to any person whatsoever, shall be guilty of a misdemeanour, and being convicted thereof shall be liable, at the discretion of the court, to be imprisoned for any term not exceeding two years.”

And here are the most reasonable views of the former Olympian Chris Boardman on this affair and the hysteria it created…

a conceptual introduction to HMC [reply from the author]

Posted in Statistics with tags , , , , , , , , on September 8, 2017 by xi'an

[Here is the reply on my post from Michael Bétancourt, detailed enough to be promoted from comment to post!]

As Dan notes this is meant as an introduction for those without a strong mathematical background, hence the focus on concepts rather than theorems! There’s plenty of maths deeper in the references. ;-)

 I am not sure I get this sentence. Either it means that an expectation remains invariant under reparameterisation. Or something else and more profound that eludes me. In particular because Michael repeats later (p.25) that the canonical density does not depend on the parameterisation.

What I was trying to get at is that expectations and really all of measure theory are reparameteriztion invariant, but implementations of statistical algorithms that depend on parameterization-dependent representations, namely densities, are not. If your algorithm is sensitive to these parameterization dependencies then you end up with a tuning problem — which parameterization is best? — which makes it harder to utilize the algorithm in practice.

Exact implementations of HMC (i.e. without an integrator) are fully geometric and do not depend on any chosen parameterization, hence the canonical density and more importantly the Hamiltonian being an invariant objects. That said, there are some choices to be made in that construction, and those choices often look like parameter dependencies. See below!

“Every choice of kinetic energy and integration time yields a new Hamiltonian transition that will interact differently with a given target distribution (…) when poorly-chosen, however, the performance can suffer dramatically.”

This is exactly where it’s easy to get confused with what’s invariant and what’s not!

The target density gives rise to a potential energy, and the chosen density over momenta gives rise to a kinetic energy. The two energies transform in opposite ways under a reparameterization so their sum, the Hamiltonian, is invariant.

Really there’s a fully invariant, measure-theoretic construction where you use the target measure directly and add a “cotangent disintegration”.

In practice, however, we often choose a default kinetic energy, i.e. a log density, based on the parameterization of the target parameter space, for example an “identify mass matrix” kinetic energy. In other words, the algorithm itself is invariant but by selecting the algorithmic degrees of freedom based on the parameterization of the target parameter space we induce an implicit parameter dependence.

This all gets more complicated when we introducing the adaptation we use in Stan, which sets the elements of the mass matrix to marginal variances which means that the adapted algorithm is invariant to marginal transformations but not joint ones…

The explanation of the HMC move as a combination of uniform moves along isoclines of fixed energy level and of jumps between energy levels does not seem to translate into practical implementations, at least not as explained in the paper. Simulating directly the energy distribution for a complex target distribution does not seem more feasible than moving up likelihood levels in nested sampling.

Indeed, being able to simulate exactly from the energy distribution, which is equivalent to being able to quantify the density of states in statistical mechanics, is intractable for the same reason that marginal likelihoods are intractable. Which is a shame, because conditioned on those samples HMC could be made embarrassingly parallel!

Instead we draw correlated samples using momenta resamplings between each trajectory. As Dan noted this provides some intuition about Stan (it reduced random walk behavior to one dimension) but also motivates some powerful energy-based diagnostics that immediately indicate when the momentum resampling is limiting performance and we need to improve it by, say, changing the kinetic energy. Or per my previous comment, by keeping the kinetic energy the same but changing the parameterization of the target parameter space. :-)

In the end I cannot but agree with the concluding statement that the geometry of the target distribution holds the key to devising more efficient Monte Carlo methods.

Yes! That’s all I really want statisticians to take away from the paper. :-)

Das Kapital [not a book review]

Posted in Statistics with tags , , , , , , , , , , , on August 18, 2017 by xi'an

A rather bland article by Gareth Stedman Jones in Nature reminded me that the first volume of Karl Marx’ Das Kapital is 150 years old this year. Which makes it appear quite close in historical terms [just before the Franco-German war of 1870] and rather remote in scientific terms. I remember going painstakingly through the books in 1982 and 1983, mostly during weekly train trips between Paris and Caen, and not getting much out of it! Even with the help of a cartoon introduction I had received as a 1982 Xmas gift! I had no difficulty in reading the text per se, as opposed to my attempt of Kant’s Critique of Pure Reason the previous summer [along with the other attempt to windsurf!], as the discourse was definitely grounded in economics and not in philosophy. But the heavy prose did not deliver a convincing theory of the evolution of capitalism [and of its ineluctable demise]. While the fundamental argument of workers’ labour being an essential balance to investors’ capital for profitable production was clearly if extensively stated, the extrapolations on diminishing profits associated with decreasing labour input [and the resulting collapse] were murkier and sounded more ideological than scientific. Not that I claim any competence in the matter: my attempts at getting the concepts behind Marxist economics stopped at this point and I have not been seriously thinking about it since! But it still seems to me that the theory did age very well, missing the increasing power of financial agents in running companies. And of course [unsurprisingly] the numerical revolution and its impact on the (des)organisation of work and the disintegration of proletariat as Marx envisioned it. For instance turning former workers into forced and poor entrepreneurs (Uber, anyone?!). Not that the working conditions are particularly rosy for many, from a scarsity of low-skill jobs, to a nurtured competition between workers for existing jobs (leading to extremes like the scandalous zero hour contracts!), to minimum wages turned useless by the fragmentation of the working space and the explosion of housing costs in major cities, to the hopelessness of social democracies to get back some leverage on international companies…

London calling

Posted in Statistics with tags , , , , , , on June 4, 2017 by xi'an

London calling to the imitation zone
Forget it, brother, you can go it alone
London calling to the zombies of death
Quit holding out and draw another breath
London calling and I don’t want to shout

The Clash, 1979

Russell Maliphant Company

Posted in Kids, pictures with tags , , , , on May 24, 2017 by xi'an

Last weekend, the Russell Maliphant Company from London was performing in a theatre nearby (in our backyard!) and we managed to get tickets at the last minute. While I am not at all versed in modern dance, this was a fantastic experience, with very moving performances from (guest) dancers like Lucia Lacarra and Marlon Dino above, a very elaborate impact of lighting that managed to duplicate or cancel depth, space, and time, great musical tracks, and a unique quality in the movements of the dancers.