And yet another roman noir taking place in Iceland! My bedside read over the past two months was “Someone to watch over me” by Yrsa Sigurðardóttir. (It took that long because I was mostly away in July and August, not because the book was boring me to sleep every night!) It is a fairly unusual book in several respects: the setting is an institution for mentally handicapped patients that was set on fire, killing five of the patients as a result, the investigator is an Icelandic lawyer, Þóra Guðmundsdóttir, along with her German unemployed-banker boyfriend, the action takes place at the height [or bottom!] of the Icelandic [and beyond!] economic crisis, when most divorce settlements are about splitting the debts of the household, and when replacing a computer becomes an issue, some of the protagonists, including the main suspects, are mentally ill, and the police and justice are strangely absent from most of the story. The the book tells a lot about the Icelandic society, where a hit-and-run is so unheard of that the police is clueless. Or seems to be. And where people see ghosts. Or think they do, as the author plays (heavily?) on the uncertainty about those ghosts. (At least, there are no elves. Nor trolls.) Definitely more in tune with the “true” Iceland than Available dark. (Well, as far as I can tell!) The mystery itself is a wee bit stretched and the final resolution slightly disappointing, implying some unlikely behaviour from the major characters. In particular, I do not buy the explanation motivating the arson itself. Terrible cover too. And not a great title in English (Watch me or Look at me would have been better) given the many books, movies and songs with the same title. Nonetheless, I liked very much the overall atmosphere of the book, enough to recommend it.
Archive for Reykjavik
As usual, Indriðason’s books are more about the past (of characters as well as of the whole country) than about current times. Voices does not switch from this pattern, the more because it is one of the earliest Inspector Erlendur’s books. Besides the murder of an hotel employee at the fringe of homelessness, lies the almost constant questioning in Indriðason’s books of the difficult or even impossible relations between parents and children and/or between siblings, and of the long-lasting consequences of this generation gap. The murder iitself is but a pretext to investigations on that theme and the murder resolution is far from the central point of the book. The story itself is thus less compelling than others I have read, maybe because the main character spends so much time closeted in his hotel room. But it nonetheless fits well within the Erlendur series. And although it is unrelated with the story, the cover reminded me very much of the Gullfoss waterfalls.
The second book, Strange Shores, is the farthest to a detective stories in the whole series. Indeed, Erlendur is back to his childhood cottage in Eastern Iceland, looking for a resolution of his childhood trauma, loosing his younger brother during a snowstorm. He also investigates another snowstorm disappearance, interrogating the few survivors and reluctant witnesses from that time. Outside any legal mandate. Sometimes very much outside! While the story is not completely plausible, both in the present and in the past, it remains a striking novel, even on its own. (Although it could read better after the earlier novels in the series.) Not only the resolution of the additional disappearance brings additional pain and no comfort to those involved, but the ending of Erlendur’s own quest is quite ambiguous. As the book reaches its final pages, I could not decide if he had reached redemption and deliverance and the potential to save his own children, or he was beyond redemption, reaching another circle of Hell. As explained by the author in an interview, this is intentional and not not the consequence of my poor understanding: ” Readers of Strange Shores are not quite certain what to make of the ending regarding Erlendur, and I’m quite happy to leave them in the dark!”. If the main character of this series focussing more on missing persons than on detective work, what’s next?!
In a somewhat desperate rush (started upon my return from Iceland and terminated on my return from Edinburgh), Marco Banterle, Clara Grazian and I managed to complete and submit our paper by last Friday evening… It is now arXived as well. The full title of the paper is Accelerating Metropolis-Hastings algorithms: Delayed acceptance with prefetching and the idea behind the generic acceleration is (a) to divide the acceptance step into parts, towards a major reduction in computing time that outranks the corresponding reduction in acceptance probability and (b) to exploit this division to build a dynamic prefetching algorithm. The division is to break the prior x likelihood target into a product such that some terms are much cheaper than others. Or equivalently to represent the acceptance-rejection ratio in the Metropolis-Hastings algorithm as
again with significant differences in the computing cost of those terms. Indeed, this division can be exploited by checking for each term sequentially, in the sense that the overall acceptance probability
is associated with the right (posterior) target! This lemma can be directly checked via the detailed balance condition, but it is also a consequence of a 2005 paper by Andrès Christen and Colin Fox on using approximate transition densities (with the same idea of gaining time: in case of an early rejection, the exact target needs not be computed). While the purpose of the recent [commented] paper by Doucet et al. is fundamentally orthogonal to ours, a special case of this decomposition of the acceptance step in the Metropolis–Hastings algorithm can be found therein. The division of the likelihood into parts also allows for a precomputation of the target solely based on a subsample, hence gaining time and allowing for a natural prefetching version, following recent developments in this direction. (Discussed on the ‘Og.) We study the novel method within two realistic environments, the first one made of logistic regression targets using benchmarks found in the earlier prefetching literature and a second one handling an original analysis of a parametric mixture model via genuine Jeffreys priors. [As I made preliminary notes along those weeks using the 'Og as a notebook, several posts on the coming days will elaborate on the above.]
Today in Warwick, I had a very nice discussion with Michael Betancourt on many statistical and computational issues but at one point in the conversation we came upon the trouble of bridging the gap between the machine learning and statistics communities. While a conference like AISTATS is certainly contributing to this, it does not reach the main bulk of the statistics community. Since, in Reykjavik, we had discussed the corresponding difficulty of people publishing a longer and “more” statistical paper in a “more” statistical journal, once the central idea was published in a machine learning conference proceeding like NIPS or AISTATS. we had this idea that creating a special fast-track in a mainstream statistics journal for a subset of those papers, using for instance a tailor-made committee in that original conference, or creating an annual survey of the top machine learning conference proceedings rewritten in a more” statistical way (and once again selected by an ad hoc committee) would help, at not too much of a cost for inducing machine learners to make the extra-effort of switching to another style. From there, we enlarged the suggestion to enlist a sufficient number of (diverse) bloggers in each major conference towards producing quick but sufficiently informative entries on their epiphany talks (if any), possibly supported by the conference organisers or the sponsoring societies. (I am always happy to welcome any guest blogger in conferences I attend!)
It took me a fairly long while to realise there was a map of Iceland as a tag-cloud at the back of the AISTATS 2014 tee-shirt! As it was far too large for me, I thought about leaving it at the conference desk last week. I did bring it back for someone the proper size though and discovered the above when unfolding the tee… Nice but still not my size!