Archive for Vancouver

Raven and the first men [jatp]

Posted in pictures, Travel with tags , , , , , , , , , , , , on August 16, 2018 by xi'an

Squamish snapshot [jatp]

Posted in Mountains, pictures, Travel with tags , , , , , , , , , , on August 14, 2018 by xi'an

Canadian Rockies [jatp]

Posted in Mountains, pictures, Travel with tags , , , , , , , on August 11, 2018 by xi'an

JSM 2018 [#4½]

Posted in Statistics, University life with tags , , , , , , , , on August 10, 2018 by xi'an

As I wrote my previous blog entry on JSM2018 before the sessions, I did not have the chance to comment on our mixture session, which I found most interesting!, with new entries on the topic and a great discussion by Bettina Grün. Including the important call for linking weights with the other parameters, as both groups being independent does not make sense when the number of components is uncertain. (Incidentally our paper with Kaniav kamary and Kate Lee does create a dependence.) The talk by Deborah Kunkel was about anchored mixture estimation, a joint work with Mario Peruggia, another arXival that I had missed.

The notion of anchoring found in this paper is to allocate specific observations to specific components. These observations are thus anchored to these components. Among other things, this modification of the sampling model implies a removal of the unidentifiability problem. Hence formally of the label-switching or lack thereof issue. (Although, as Peter Green repeatedly mentioned, visualising the parameter space as a point process eliminates the issue.) This idea is somewhat connected with the constraint Jean Diebolt and I imposed in our 1990 mixture paper, namely that no component would have less than two observations allocated to it, but imposing which ones are which of course reduces drastically the complexity of the model. Another (related) aspect of anchoring is that the observations that are anchored to the components act as parts of the prior model, modifying the initial priors (which can then become improper as in our 1990 paper). The difficulty of the anchoring approach is to find observations to anchor in an unsupervised setting. The paper proceeds by optimising the allocations, which somewhat turns the prior into a data-dependent prior since all observations are used to set the anchors and then used again for the standard Bayesian processing. In that respect, I would rather follow the sequential procedure developed by Nicolas Chopin and Florian Pelgrin, where the number of components grows by steps with the number of observations.

 

Locarno beach [jatp]

Posted in Mountains, pictures, Travel with tags , , , , , , on August 8, 2018 by xi'an

Stawamus Chief

Posted in Mountains, pictures, Travel with tags , , , , , , , on August 6, 2018 by xi'an

Being back in Vancouver gave me the great opportunity to bag a climb I had wanted to make for quite a while, the classical route on the Stawamus (or Squamish) Chief, which is a big granite dome standing on top of the Howe Sound, north of Vancouver and south of Whistler (for NIPS longterm attendees!). In 2011, Julien, David and I went climbing for a whole day a cliff nearby called Burger and Fries. While the Chief is a 600m big vertical wall that makes for a fairly involved climb, the standard route is mostly bypassing the vertical exposure and goes up on gritty granite slabs that do not require hand pulls (The Apron) and then cracks that make for an easy climb, until the crux of the climb (Buff), again involving cracks if vertical ones and a half chimney, somewhat more exposed, with a solo free climber passing my guide Brett Nixon from Vancouver Mountain Guides and myself on the way. Something around 12 pitches total. Just a great climb, well-suited for my lack of sufficient training in the past months!, and with terrific views all the way, plus constant shade a big plus on a very hot day! Meeting with friends at the top, who had gone up the hiking path, was an added bonus as we could hike down together. Hopefully, I will be back in a near future to try another route, like Angel’s Crest…

for a coincidence

Posted in Mountains, pictures, Travel, University life with tags , , , , , , on August 5, 2018 by xi'an

Last night in Vancouver, we were walking back to Chinatown under an expressway, in a rather uninspiring section of town. Waiting at a cross-light with another couple on the other side. As we crossed the street I glanced at the man and noticed his Chamonix North Face tee-shirt. He happened to do the same and… noticed my identical Chamonix North Face tee-shirt! We shared a laugh at this (huge?) coincidence and continued on our respective ways. (He was not taking part in MCMskiii in case this seems a likely explanation!)