Personal website http://chauff.github.io Update: a slide deck is now available as well - a presentation about IR in general and the axiomatic approach to neural IR models, given to MSc students as part of the Delft Deep Learning Colloqium 2019. Introduction Traditional retrieval models such as BM25 or language models have been engineered... Tue, 16 Apr 2019 00:00:00 +0200 http://chauff.github.io/2019-04-16-axiomatic/ http://chauff.github.io/2019-04-16-axiomatic/ Browsing through the accepted papers of ICLR 2019 (the conference takes place in May 2019, the notifications came out around Christmas 2018) a few works stood out to me as an IR researcher - the list is below. Compared to the previous years, my impression is that the number of... Thu, 07 Feb 2019 00:00:00 +0100 http://chauff.github.io/2019-02-07-iclr2019/ http://chauff.github.io/2019-02-07-iclr2019/ While today’s Web search engines are designed for single-user search, over the years research efforts have shown that complex information needs - which are explorative, open-ended and multi-faceted - can be answered more efficiently and effectively when searching in collaboration. Collaborative search (and sensemaking) research has investigated techniques, algorithms and... Wed, 12 Dec 2018 00:00:00 +0100 http://chauff.github.io/2018-12-12-groupsize/ http://chauff.github.io/2018-12-12-groupsize/ Once a year, I try to attend an NLP or ML conference, usually picking the one that is closest in terms of location. In 2016 I was at ACL in Berlin :de:, in 2017 I attended ICLR in Toulon :fr: and this year I made the trip to EMNLP in... Sun, 04 Nov 2018 00:00:00 +0100 http://chauff.github.io/2018-11-04-emnlp/ http://chauff.github.io/2018-11-04-emnlp/ As part of the NWA FairNews project we have recently been working on the question of how to provide fair access to online news. And of course that question itself leads to a whole lot of other questions, such as when is something actually fair or biased or … and... Mon, 15 Oct 2018 00:00:00 +0200 http://chauff.github.io/2018-10-15-siren/ http://chauff.github.io/2018-10-15-siren/ In recent months, we spent a lot of time and effort to design and implement a well-functioning open-source collaborative search system called SearchX. I wrote about it here. Having done all the implementation work, now is the time to reap the benefits of the work, i.e. run experiments with our... Tue, 07 Aug 2018 00:00:00 +0200 http://chauff.github.io/2018-08-07-sal-at-cikm/ http://chauff.github.io/2018-08-07-sal-at-cikm/ Web search is generally seen as a solitary activity, as most mainstream technologies are designed for single-user search sessions. However, for a sufficiently complex task, search collaborations have shown to be beneficial. Existing systems In one of my projects, large-scale collaborations whilst searching (think tens or hundreds of potential collaborators)... Sat, 21 Jul 2018 00:00:00 +0200 http://chauff.github.io/2018-07-21-collaborative-search/ http://chauff.github.io/2018-07-21-collaborative-search/ After a few years of conducting experiments in the MOOC environment and often finding that we were not able to replicate major insights from prior studies (conducted in the lab, classroom, a specifically designed online learning environment), we had an inkling that MOOC environments are tougher to crack. In order... Tue, 05 Jun 2018 00:00:00 +0200 http://chauff.github.io/2018-06-05-metareview-active-learning/ http://chauff.github.io/2018-06-05-metareview-active-learning/ UPDATE: the positions are no longer available. In the NWO project Searchx: Integrating search and sensemaking into large-scale open online learning I have two fully-funded 4-year PhD positions at the Delft University of Technology. Application deadline: June 15, 2018 (the advert will remain accessible until the positions are filled). Project... Mon, 28 May 2018 00:00:00 +0200 http://chauff.github.io/2018-05-28-phd-positions/ http://chauff.github.io/2018-05-28-phd-positions/ When I am not talking about information retrieval research/education, I usually talk about MOOCs in this blog. While IR and MOOCs seem pretty unrelated, they come together very well in the area of Search As Learning. Search though is not always necessary for MOOC research, there are plenty of other... Thu, 03 May 2018 00:00:00 +0200 http://chauff.github.io/2018-05-03-intellieye/ http://chauff.github.io/2018-05-03-intellieye/ Imagine this: you are an educator who teaches a large class and would like to give your students the opportunity to practice as much as they like. So you create questions. But that is cumbersome. Instead, you’d like (or: I’d like) an automatic pipeline that takes in the material you... Mon, 30 Apr 2018 00:00:00 +0200 http://chauff.github.io/2018-04-30-learningq/ http://chauff.github.io/2018-04-30-learningq/ After quite a few years of not teaching information retrieval, I finally taught an IR MSc course (140 hours study load, 2 lectures a week for 8 weeks; 4 support hours per week) together with my colleague Nava Tintarev earlier this year. While Nava introduced the students to the basics... Tue, 24 Apr 2018 00:00:00 +0200 http://chauff.github.io/2018-04-24-ir/ http://chauff.github.io/2018-04-24-ir/ Schloss Dagstuhl is a familiar name for many Computer Science researchers. A center founded in the early 1990s that has quickly become one of the world’s premier meeting centers for computer science research. It is also in the middle of nowhere in Germany, which is a pain in terms of... Fri, 15 Sep 2017 00:00:00 +0200 http://chauff.github.io/2017-09-15-sal/ http://chauff.github.io/2017-09-15-sal/ It is conference season. Again. Not just in the IR, but also the NLP and ML communities. Time to start reading. ACL 2017 with its nearly 200 full papers and more than 100 short papers was a treasure trove again for inspiring approaches, problems and datasets. There is really only... Tue, 22 Aug 2017 00:00:00 +0200 http://chauff.github.io/2017-08-22-acl2017/ http://chauff.github.io/2017-08-22-acl2017/ NWO VIDI Awarded The Netherlands Organisation for Scientific Research has just announced the proposals that will receive funding through the NWO VIDI scheme. I made it onto this list! The VIDI is a personal grant worth 800K € which translates into 2 PhD and 1 postdoc positions and more research... Tue, 30 May 2017 00:00:00 +0200 http://chauff.github.io/2017-05-30-vidi/ http://chauff.github.io/2017-05-30-vidi/ I read papers. A lot of them. Most of them conference papers. SIGIR, ACL, EMNLP, CIKM, WWW, WSDM, CHI, … I try to at least browse through the published abstracts every year. Some of those papers stick with me for a long time, because the idea is quirky, the approach... Tue, 02 May 2017 00:00:00 +0200 http://chauff.github.io/2017-05-02-glowbox/ http://chauff.github.io/2017-05-02-glowbox/ Deep learning has been the dominant theme in a number of research communities (e.g. computer vision, natural language processing) for the past few years. In information retrieval (IR), things are moving more slowly, we are still experimenting with word embeddings and are only beginning to adapt the (by now) classic... Tue, 25 Apr 2017 00:00:00 +0200 http://chauff.github.io/2017-04-25-iclr2017/ http://chauff.github.io/2017-04-25-iclr2017/ I have just finished this academic year’s edition of Big Data Processing, a Bachelor course I have been teaching for a few years now. As I am not a big fan of teaching the same course for 5+ years, I am happy to move on to a different course next... Tue, 07 Feb 2017 00:00:00 +0100 http://chauff.github.io/2017-02-07-quizzes/ http://chauff.github.io/2017-02-07-quizzes/ Top Grant The Netherlands Organisation for Scientific Research has recently announced the list of proposals awarded funding through the Top Grant Scheme. The TOP2 is a personal grant worth about 220K € which funds a 4-year PhD position and some extras. This year my proposal on Large-Scale Collaborative Search &... Sun, 27 Nov 2016 00:00:00 +0100 http://chauff.github.io/2016-11-27-lacrosse/ http://chauff.github.io/2016-11-27-lacrosse/ MOOCs aim to “educate the world.” More often than not, however, MOOCs fall short of this goal — a majority of learners are already highly educated and come from specific parts of the (developed) world. Learners from developing countries without a higher degree are underrepresented, though desired, in MOOCs. One... Tue, 27 Sep 2016 00:00:00 +0200 http://chauff.github.io/2016-09-27-from-learners-to-earners/ http://chauff.github.io/2016-09-27-from-learners-to-earners/