the alpinist [film review]

Watched (with supplementary oxygen) The Alpinist in the plane to Jeddah. It is a documentary (made by the same filmmakers who filmed the Dawn Wall) about the amazing Canadian alpinist Marc-André Leclerc, who died in 2018 on the Mendenhall Glacier, Alaska, in an avalanche, after achieving extraordinary complex solo climbs as eg on Mount Robbson, Cerro Torre in Patagonia. These are winter climbs, partly ice climbing, where no repetition is possible and where the objective conditions (hence dangers) may vary considerably. In that regard, these achievements could be argued to go even beyond Alex Honnold’s free solo climb of El Capitan, where Honnold practiced the route over and over before making his successful free solo attempt. (Obviously an inhuman achievement when considering the hardest bits are at least 7c!) Watching Marc-André Leclerc when mixed climbing is just as heart stopping as watching Honnold rock climbing. He must have been incredibly strong to master these monstruous icy walls and maintain his absolute vigilance in each crampon move, in each ice-pick placement. Sadly it ended up with an avalanche… I obviously enjoyed X’ing many places I had visited, like the approach walk to Robson (in 1991!), the Three Sisters of Canmore [below], and routes of Squamish [above]. (I wonder who filmed during these non-advertised climbs. For instance, he told no one except his partner when he summited Mount Robson. In some cases he was clearly self-filing at lower intensity points, but in others it could have been an helicopter (or a drone?). In this respect, but by far not only in this respect, his blog is definitely worth the read.)

~

position at UBC

The Department of Statistics at the University of British Columbia, Vancouver invites applications from outstanding new investigators for a tenure-track position at the rank of Assistant Professor in the area of statistical machine learning and artificial intelligence (AI), with an anticipated start date of July 1, 2022 or January 1, 2023.

This position is offered as part of a new, interdisciplinary research cluster, AI Methods for Scientific Impact (AIM-SI) within UBC’s Centre for AI Decision-making and Action (CAIDA). CAIDA consists of over 100 researchers whose research leverages AI; AIM-SI will hire 5 new faculty members across three departments: Computer Science, Mathematics, and Statistics. Those recruited will join over a dozen existing researchers who are highly active in the core AI Methods research community.

Candidates must have a Ph.D. in Statistics, Biostatistics, or a related field and must demonstrate evidence of research success and a high potential to be leaders in their research field. The successful candidate should have a strong record of research productivity commensurate with their experience and will be expected to develop an independent research program in an area that complements existing Department expertise and aligns with opportunities at UBC. The successful candidate will be expected to effectively supervise Statistics graduate students, collaborate with other faculty members, obtain external funding, teach undergraduate and graduate Statistics courses, and actively participate in departmental activities.

Equity and diversity are essential to academic excellence. An open and diverse community fosters the inclusion of voices that have been underrepresented or discouraged. We encourage applications from members of groups that have been marginalized on any grounds enumerated under the B.C. Human Rights Code, including sex, sexual orientation, gender identity or expression, racialization, disability, political belief, religion, marital or family status, age, and/or status as a First Nation, Métis, Inuit, or Indigenous person.

Applicants will be asked to complete an equity survey. The survey information will not be used to determine eligibility for employment, but will be collated to provide data that can assist us in understanding the diversity of our applicant pool and identifying potential barriers to the employment of designated equity group members. Applicants’ participation in the survey is voluntary and confidential and takes only a minute to complete. Applicants may self-identify in one or more of the designated equity groups, or may also decline to identify in any or all of the questions by choosing “not disclosed.”

All qualified candidates are encouraged to apply; however, Canadians and permanent residents will be given priority.

Deadline for application is December 15. More information is available at

https://www.stat.ubc.ca/assistant-professor-aim-si-tenure-track-position

ISBA 2021 grand finale

Last day of ISBA (and ISB@CIRM), or maybe half-day, since there are only five groups of sessions we can attend in Mediterranean time.

My first session was one on priors for mixtures, with 162⁺ attendees at 5:15am! (well, at 11:15 Wien or Marseille time), Gertrud Malsiner-Walli distinguishing between priors on number of components [in the model] vs number of clusters [in the data], with a minor question of mine whether or not a “prior” is appropriate for a data-dependent quantity. And Deborah Dunkel presenting [very early in the US!] anchor models for fighting label switching, which reminded me of the talk she gave at the mixture session of JSM 2018 in Vancouver. (With extensions to consistency and mixtures of regression.) And Clara Grazian debating on objective priors for the number of components in a mixture [in the Sydney evening], using loss functions to build these. Overall it seems there were many talks on mixtures and clustering this year.

After the lunch break, when several ISB@CIRM were about to leave, we ran the Objective Bayes contributed session, which actually included several Stein-like minimaxity talks. Plus one by Théo Moins from the patio of CIRM, with ciccadas in the background. Incredibly chaired by my friend Gonzalo, who had a question at the ready for each and every speaker! And then the Savage Awards II session. Which ceremony is postponed till Montréal next year. And which nominees are uniformly impressive!!! The winner will only be announced in September, via the ISBA Bulletin. Missing the ISBA general assembly for a dinner in Cassis. And being back for the Bayesian optimisation session.

I would have expected more talks at the boundary of BS & ML (as well as COVID and epidemic decision making), the dearth of which should be a cause for concern if researchers at this boundary do not prioritise ISBA meetings over more generic meetings like NeurIPS… (An exception was George Papamakarios’ talk on variational autoencoders in the Savage Awards II session.)

Many many thanks to the group of students at UConn involved in setting most of the Whova site and running the support throughout the conference. It indeed went on very smoothly and provided a worthwhile substitute for the 100% on-site version. Actually, I both hope for the COVID pandemic (or at least the restrictions attached to it) to abate and for the hybrid structure of meetings to stay, along with the multiplication of mirror workshops. Being together is essential to the DNA of conferences, but travelling to a single location is not so desirable, for many reasons. Looking for ISBA 2022, a year from now, either in Montréal, Québec, or in one of the mirror sites!

parallel tempering on optimised paths

Saifuddin Syed, Vittorio Romaniello, Trevor Campbell, and Alexandre Bouchard-Côté, whom I met and discussed with on my “last” trip to UBC, on December 2019, just arXived a paper on parallel tempering (PT), making the choice of tempering path an optimisation problem. They address the touchy issue of designing a sequence of tempered targets when the starting distribution π⁰, eg the prior, and the final distribution π¹, eg the posterior, are hugely different, eg almost singular.

“…theoretical analysis of reversible variants of PT has shown that adding too many intermediate chains can actually deteriorate performance (…) [while] on non reversible regime adding more chains is guaranteed to improve performances.”

The above applies to geometric combinations of π⁰ and π¹. Which “suffers from an arbitrarily suboptimal global communication barrier“, according to the authors (although the counterexample is not completely convincing since π⁰ and π¹ share the same variance). They propose a more non-linear form of tempering with constraints on the dependence of the powers on the temperature t∈(0,1).  Defining the global communication barrier as an average over temperatures of the rejection rate, the path characteristics (e.g., the coefficients of a spline function) can then be optimised in terms of this objective. And the temperature schedule is derived from the fact that the non-asymptotic round trip rate is maximized when the rejection rates are all equal. (As a side item, the technique exposed in the earlier tempering paper by Syed et al. was recently exploited for a night high resolution imaging of a black hole from the M87 galaxy.)

Thin Air [book review]

When visiting Vancouver last December [at a time when traveling was still possible], I had the opportunity to revisit White Dwarf Books [thirty years after my first visit] and among other books bought a Richard Morgan‘s novel, Thin Air, that I did not know existed and which was recommended by the [friendly] book seller. As superior to Morgan’s foray into dark fantasy (that I did not dislike so much). As I had really enjoyed the Altered Carbon series, I jumped on this new novel, which is a form of sequel to Th1rt3en, and very very similar in its futuristic pastiche of tough detectives à la Marlowe, dry humour included. A form of space noir, as The Guardian puts it. I sort of got quickly lost in the plot and (unusually) could not keep track of some characters, which made reading the book a chore towards the end. Thanks to the COVID-19 quarantine, I still managed to finish it, while home cycling!, the very end being more exciting than the beginning drudgery and the predictable sex scenes bound to occur in every of his novels. The Martian world in the novel is only alluded to, which makes it more appealing, despite the invasive jargon, however it sounds too much like a copy of our 20th century with car chase and gun/knife fights. Enhanced by an embedded AI when one can afford it. Certainly not the best read in the series but enough to tempt me into looking at the first episodes of Altered Carbon on Netflix. [Note: the book is not to be confused with the bestselling Krakauer’s Into Thin Air, which relates the 1996 Everest disaster, soon turned into a rather poor movie. I had not realised till today that the same Krakauer wrote Into the Wild…!]