I’ve been falling behind with these news updates, so here is a summary over the last six months.
I got a few papers accepted for publication at/in ICLR (Calibration tests beyond classification with David Widmann and Dave Zachariah), IJCAI (Likelihood-free Out-of-Distribution Detection with Invertible Generative Models with Amir Ahmadian), IEEE IV Workshop on Data Driven Intelligent Vehicle Applications (Self-Supervised Representation Learning for Content Based Image Retrieval of Complex Scenes with Hari Govindarajan, Peter Lindskog, Dennis Lundström, Amanda Olmin, and Jacob Roll), IEEE Control Systems Magazine (Nonlinear System Identification - Learning while respecting physical models using Sequential Monte Carlo with Anna Wigren, Johan Wågberg, Adrian Wills and Thomas Schön), and JMLR (Pseudo-Marginal Hamiltonian Monte Carlo with Johan Alenlöv and Arnaud Doucet). The latter paper only took 5 years from initial submission to acceptance :) A lot has happended with the paper in this period of time, though, so if you happened to read it back in 2016 and found the theoretical analysis lacking, I recommend having a look at the latest version.
Our introductory text book, Machine Learning - A First Course for Engineers and Scientists with Andreas Lindholm, Niklas Wahlström, and Thomas Schön, has finally been passed on to Cambridge University Press. I’m looking forward to seeing the printed version of the book!
During the spring we introduced a Linköping University version of the online self-study course Elements of AI, Part 2: Building AI. To all Swedes who have taken the course, feel free to sign up for our version to get university credits for it. On a similar note, in September we will launch a newly developed online course on the Foundations of Machine Learning, based on the aforementioned book.
That’s all for now. During the autumn I will be on parental leave, looking after our youngest daughter, so it’s quite likely that it will be another six months until the next update.