Archive for Cambodia
horsy new year!
Posted in pictures, Travel with tags ⾺, Cambodia, China, Chinese zodiac, Fire Horse, horse, Japan, Lunar New Year, lunisolar calendar, North Korea, Singapore, South Korea, Year of the Horse on February 17, 2026 by xi'anAngkor beer
Posted in pictures, Travel, Wines with tags Angkor beer, beer, Cambodia, lager, Sianoukville, Siem Reap on July 20, 2019 by xi'anLa peste et la vigne [book review]
Posted in Books, Kids, Travel with tags book review, Cambodia, fantasy, French book, Joe Abercrombie, mercenary company, plague, Robin Hobb on March 17, 2019 by xi'an
During my trip to Cambodia, I read the second volume of this fantasy cycle in French. Which I liked almost as much as the first volume since the author continues to explore the mystery of the central character Syffe and its relations with some magical forces at play in his universe. As in most stories uniquely centred on a single character point of view the recurring ponderings of Syffe about his role in life, the existence of supernatural forces, and his own sanity may tend to get annoying at time. But the escape from the mines and the subsequent stay in a mountain kingdom are well-paced, especially the description of the plague that allows such an escape. The last section is more connected with the first volume and sees more warfare, again with sudden reversals of fortune (no further spoiler!). The final chapters see a lot explained about many aspects of the story and the raison d’être of the character, even though the very last surprise is somewhat predictable. But opening new vistas for the future volumes. There are still many threads I could have pulled to point some potential influences of earlier cycles, from Stephen Donaldson’s Thomas Covenant chronicles, which I simply hated!, to Robin Hobb’s Soldier’s son. Since both stories convey the feeling of a magical force at the level of the whole land (or universe), with the unprepared and imperfect “hero” able to impact this land in dramatic ways. And again Elizabeth Moon’s Deeds of Paksenarion for the depiction of mercenary companies…
optimal choice among MCMC kernels
Posted in Statistics with tags Angkor Wat, Cambodia, delayed acceptance, filamentary distribution, invariance, invariant measure, Markov kernel, Normandie, population Monte Carlo, Siem Reap, sparsity on March 14, 2019 by xi'an
Last week in Siem Reap, Florian Maire [who I discovered originates from a Norman town less than 10km from my hometown!] presented an arXived joint work with Pierre Vandekerkhove at the Data Science & Finance conference in Cambodia that considers the following problem: Given a large collection of MCMC kernels, how to pick the best one and how to define what best means. Going by mixtures is a default exploration of the collection, as shown in (Tierney) 1994 for instance since this improves on both kernels (esp. when each kernel is not irreducible on its own!). This paper considers a move to local weights in the mixture, weights that are not estimated from earlier simulations, contrary to what I first understood.
As made clearer in the paper the focus is on filamentary distributions that are concentrated nearby lower-dimension sets or manifolds Since then the components of the kernel collections can be restricted to directions of these manifolds… Including an interesting case of a 2-D highly peaked target where converging means mostly simulating in x¹ and covering the target means mostly simulating in x². Exhibiting a schizophrenic tension between the two goals. Weight locally dependent means correction by Metropolis step, with cost O(n). What of Rao-Blackwellisation of these mixture weights, from weight x transition to full mixture, as in our PMC paper? Unclear to me as well [during the talk] is the use in the mixture of basic Metropolis kernels, which are not absolutely continuous, because of the Dirac mass component. But this is clarified by Section 5 in the paper. A surprising result from the paper (Corollary 1) is that the use of local weights ω(i,x) that depend on the current value of the chain does jeopardize the stationary measure π(.) of the mixture chain. Which may be due to the fact that all components of the mixture are already π-invariant. Or that the index of the kernel constitutes an auxiliary (if ancillary) variate. (Algorithm 1 in the paper reminds me of delayed acceptance. Making me wonder if computing time should be accounted for.) A final question I briefly discussed with Florian is the extension to weights that are automatically constructed from the simulations and the target.

Read the delightful

