A Summary of 2020

Since this is going to be my last blog post for this year, I’m going to use it for reflecting a bit on my academic life this year and for talking a bit about my plans for 2021.

Research Activities

So first of all, what has happened research-wise this year? Well, most of my research activities have focused on analyzing data from a psychological study on shapes, as discussed in various blog posts (here, here, here, and here). This cooperation with Margit Scheibel has already started in 2018 and I’m really glad that we were able to finish our analyses this summer. We’re currently in the process of writing up an article on that topic that will be submitted to a journal hopefully soon.

In the past few days, I’ve started to think about a machine learning study which is supposed to re-apply the techniques from the NOUN study (see here and here) and to look further into more complex architectures for learning a mapping from image input to coordinates in a similarity space. As targets, I’m going to use the conceptual spaces extracted from the Shapes study. They focus on a single visual domain whose structure is still poorly understood, which seems to be a very good fit for the proposed hybrid procedure.

Teaching Activities

I did not have any teaching duties in the summer term, since I was on a parental leave until June. In the current winter term, I’m holding a seminar about natural language processing. Teaching in a completely virtual setting is certainly challenging, but I’ve tried to arrange things in such a way that only very few meetings in large groups are necessary, and most of the work takes place in smaller groups. So far, things seem to go relatively smoothly and I’m looking forward to watching the students’ educational videos at the end of the semester.

In addition to the seminar, I have the pleasure of supervising two thesis projects right now: A bachelor thesis on one-shot learning in conceptual spaces and a master thesis on sentiment analysis. Both are still in a relatively early stage and will be submitted some time in 2021.

Organizational Activities

And then there was of course CARLA (in cooperation with Mingya Liu): On the one hand, we were still busy with creating an edited volume based on selected contributions to the CARLA 2018 workshop. The overall process has taken quite some time due to various kinds of delays, but in the past weeks we finally signed the contract and submitted our files. It’s probably too optimistic to hope for a publication still in 2020, but the book should hopefully be available in the first quarter of 2021.

In addition to the edited volume, we also organized another CARLA workshop this year. Originally, we had planned a physical meeting in Bolzano as part of the Bolzano Summer of Knowledge, but Corona forced us (like so many other events) to go virtual. Nevertheless, we were able to have a nice two-day event with various talks from different disciplines. A group of young researchers has volunteered to join the organizing team for the years to come and we’re currently in the process of planning CARLA 2021, probably as a hybrid event.


This year, I’ve spent quite some time working on my dissertation. Since I chose a monograph over a cumulative dissertation, I have to write some background chapters, namely, on multidimensional scaling, machine learning, and artificial neural networks. As it turns out, I have invested more time into reading literature and writing background chapters than into any other activity this year. While this progress is unfortunately not directly visible to the outside world (unlike experimental results etc.), it is nevertheless an important step for me towards obtaining my PhD and has also greatly improved my understanding of what I’m actually doing in my research.


So what are my plans for 2021?

Well, first of all, I’ll try to get the article about the psychological study on shapes out there as soon as possible. While the original plan was to submit in September 2020, it now looks more like January or February 2021.

On the research side, I’ll put my focus on the machine learning experiments I’ve started to work on with respect to the extracted shape spaces. While the overall experimental design is already relatively clear (more about that in another blog post in January), setting up the code, running the computations, and analyzing the results certainly will take some time. Nevertheless, I hope to have some reasonable results in the first half of 2021.

Other than that, my general goal for 2021 is to hand in my dissertation. I guess I could continue with my research forever, since there are so many interesting ideas to pursue, but at some point you simply have to finish up. It looks like my work on the shape domain will be my last research project for the scope of my PhD, since I will need some additional time for writing and proofreading. That unfortunately means that I probably won’t get the chance to continue with my research on Logic Tensor Networks and with the investigations regarding the rectangle domain, although both projects are still super interesting to me. But who knows, maybe I’ll be much faster than anticipated with everything else and end up having some free time? 😉

Leave a Reply