As a familial tradition of the end of the year movie, I went with my daughter to watch the second (or eighth) movie in the series. As I had heard and read several highly positive reviews on the originality of the scenario and the sharpness of the photography, I was expecting a lot from the movie. And hence was quite disappointed by the quasi-absence of scenario (never a major strength in the series anyway!) and by the pre-teens dialogues, some situations reminding me of the worst Star Trek episodes, like the very final ludicrous scene in the space shuttle… Some parts are total failures, like the expedition to the casino planet. Or the final battle scene that lasts for evvvvver… Or the initial battle scene that lasts about as long. Or the fight with the lobsters, endless! And do not even think of mentioning the Disneyian pongs. And as usual the utter disdain for any law of physics. Like a moon going from full to crescent on the same night (minor spoiler!). Terrible, all in all, except for the scenery of the Irish island, Skellig Michael, with its very primitive monastery, which reminded me of St Kilda… And a few actors surviving the disaster.
Archive for St Kilda
dimmed Star Wars
Posted in Books, Kids, Mountains, pictures, Travel with tags Disney, Ireland, movie review, Skellig Michael, St Kilda, Star Trek, Star Wars, The Last Jedi on December 25, 2017 by xi'anMelbourne coastline [jatp]
Posted in pictures, Running, Travel with tags Australia, Melbourne, sea, St Kilda, sunrise, Victoria on August 31, 2016 by xi'anlocally weighted MCMC
Posted in Books, Statistics, University life with tags Australia, effective sample size, Harvard University, Melbourne, parallel MCMC, Rao-Blackwellisation, recycling, St Kilda, vanilla Rao-Blackwellisation on July 16, 2015 by xi'anLast week, on arXiv, Espen Bernton, Shihao Yang, Yang Chen, Neil Shephard, and Jun Liu (all from Harvard) proposed a weighting scheme to associated MCMC simulations, in connection with the parallel MCMC of Ben Calderhead discussed earlier on the ‘Og. The weight attached to each proposal is either the acceptance probability itself (with the rejection probability being attached to the current value of the MCMC chain) or a renormalised version of the joint target x proposal, either forward or backward. Both solutions are unbiased in that they have the same expectation as the original MCMC average, being some sort of conditional expectation. The proof of domination in the paper builds upon Calderhead’s formalism.
This work reminded me of several reweighting proposals we made over the years, from the global Rao-Blackwellisation strategy with George Casella, to the vanilla Rao-Blackwellisation solution we wrote with Randal Douc a few years ago, both of whom also are demonstrably improving upon the standard MCMC average. By similarly recycling proposed but rejected values. Or by diminishing the variability due to the uniform draw. The slightly parallel nature of the approach also connects with our parallel MCM version with Pierre Jacob (now Harvard as well!) and Murray Smith (who now leaves in Melbourne, hence the otherwise unrelated picture).