AISTATS 2014 (day #1)

First day at AISTATS 2014! After three Icelandic vacations days driving (a lot) and hinkg (too little) around South- and West-Iceland, I joined close to 300 attendees for this edition of the AISTATS conference series. I was quite happy to be there, if only because I had missed the conference last year (in Phoenix) and did not want this to become a tradition… Second, the mix of statistics, artificial intelligence and machine learning that characterises this conference is quite exciting, if challenging at time. What I most appreciated in this discovery of the conference is the central importance of the poster session, most talks being actually introductions to or oral presentations of posters! I find this feature terrific enough (is there such a notion as “terrific enough”?!) worth adopting in future conferences I am involved in. I just wish I had managed to tour the whole collection of posters today… The (first and) plenary lecture was delivered by Peter Bühlman, who spoke about a compelling if unusual (for me) version of causal inference. This was followed by sessions on Gaussian processes, graphical models, and mixed data sources. One highlight talk was the one by Marc Deisenroth, who showed impressive robotic fast learning based on Gaussian processes. At the end of this full day, I also attended an Amazon mixer where I learned about Amazon‘s entry on the local market, where it seems the company is getting a better picture of the current and future state of the U.S. economy than governmental services, thanks to a very fine analysis of the sales and entries on Amazon‘s entry. Then it was time to bike “home” on my rental bike, in the setting sun…

AISTATS 2014 in Reykjavik, Iceland

The next AISTATS conference will take place in Reykjavik, Iceland. On April 22-25, 2014. This conference “is an interdisciplinary gathering of researchers at the intersection of computer science, artificial intelligence, machine learning, statistics, and related areas.” The deadline for paper submissions is November 1, 2013. And there is a deadline for late-breaking poster abstract submissions, namely January 24. Given my heavy travel schedule next year, I am not sure I can attend, but I am definitely tempted! Esp. since I missed AISTATS 2013 in Phoenix, where I was kindly invited, due to The Accident…

AISTAT 2013

In case you have missed the announcement, the AISTAT 2013 conference will take place in Phoenix, Arizona, on April 29-May 01, 2013.  This is the Sixteenth International Conference on Artificial Intelligence and Statistics. Registration and hotel reservation are now open. (Not that this is particularly relevant but I will attend the conference and give a lecture on, surprise, surprise!… ABC. Looking at the past location, it seems this is the first one not taking place on a beach, for which I am grateful! I am looking forward climbing near Phoenix, welcoming any suggestion to this effect.)

WSC 2012, Berlin

For once, while being away at a conference (WSC 2012), I will not be reporting on every day and session, for the simple reason I attended very few sessions and have therefore little to report! Last year in Phoenix, I had already resented a certain difficulty to relate with most of the talks and felt it unprofitable to attend most sessions, except those organised by Pierre Lécuyer and co-authors. This year, the pervasive combination of snowed-in Berlin, of a nice rental apartment (five time cheaper than the fancy Intercontinental), shared with Jean-Michel Marin whose inputs on the book were way overdue, of the distance to the Intercontinental, and of moderately exciting sessions meant that we spent most of the conference closeted in this apartment in Schöneberg, except for a quick stroll on Unter den Linden and for the inaugural conference by Stefan Rahmstorf on climate change (whose contents were surely worth broadcasting, although mildly related with simulation), a session on rare event simulation, and the session I organised on Monte Carlo Methods in Statistics (same title as the special issue of TOMACS, not coincidentally!). Actually, the feeling was mutual, apparently!, as our session attracted very few WSC 2012 attendees, a shame when considering the three great talks delivered by Anthony Lee, Jean-Michel Marin, and Nial Friel. To wit, Anthony talked about several scenarios for running ABC (see my earlier post of yesterday), Jean-Michel explained the tolerance selection developed in our recent paper, and Nial Friel discussed an on-going project dealing with composite likelihood in Gibbs random fields, a topic obviously close to my interests and the just accepted PNAS paper!

Winter simulation conference [im Berlin]

Here is a call for papers for the 2012 Winter Simulation Conference (WSC) that will take place in Berlin, December 9-12, this year, following Phoenix last year. I am organising a session there on statistical Monte Carlo methods with Nial Friel, Anthony Lee, and Jean-Michel Marin as invited speakers.

The Winter Simulation Conference (WSC) is the premier international forum for disseminating recent advances in the  field of dynamic systems modeling and simulation. In addition to a technical program of unsurpassed scope and quality, WSC provides the central meeting place for simulation practitioners, researchers, and vendors. Research in modeling and simulation is propelled by fostering cross fertilization between various disciplines. 

Regular Paper Abstracts due April 2, 2012. See http://wintersim.org/node/4 for additional deadlines and other information for Authors (including the Authors’ Kit).

Find out more about the tracks, the keynote speakers, the anniversaries that we celebrate in 2012, the venue, and how to submit a paper at: http://wintersim.org/