Design of experiments assisted by artificial intelligence
Yes, I know – I’ve touched on this topic many times before in this blog, so you have every right to be bored and surf. I have to talk about it from time to time anyway, because it’s the focus of my research these days.
Recently I was on the island of Elba (a wonderful place) for a conference on advanced detectors for fundamental physics, and I presented a poster there on the subject of artificial intelligence aided design of instruments for fundamental physics. Below is the poster (hope it’s readable in this zipped version – if you really want a better photo, ask).
After the conference, like all the other poster presenters, I was asked to produce a _two_ page article summarizing it. As I started to put it together I discovered that the fixed template they had distributed left very little room for text and figures – I found it impossible to fit in these two pages even a very short talk. I took drastic cuts and removed all references, leaving a few footnotes to save space. The result was the best I could do given the boundary conditions, and I sent it to colleagues for comments.
One of them, who was also present at the conference, then explained to me that the conference organizers had distributed two different templates: one for reviewers, in one-column format and wide-spaced text, and one for the final publication, in two-column format and much narrower fonts and spacing. The two-page limit only applied to the latter… Doh. Talk about not reading the fine print of the instruction document :-/
I will have to completely rewrite the document, of course, because in two-column format, the text I provided fits perfectly on less than one page! In the end, however, I did not waste my time: the very condensed text that I developed is in fact clearer than my usual articles. In fact, I think it deserves to be shown here. So please find it below – a very succinct description of what lies ahead in detector design!