One FLAT World Seminar


A Formal Languages and Automata Theory Seminar

After the outbreak of the coronavirus pandemic, a few scientific communities faced cancellations of conferences, seminars, and research visits. Motivated by the need to establish new communication channels, a series of seminars called One World Seminars was initiated, as an attempt to keep the communities together. The pioneer of this project was One World Probability Seminar that inspired several other One World projects (among which, you might know the One World Combinatorics on Words Seminar).

This is a new series of online research seminars on topics related to Formal Languages and Automata Theory: One FLAT World Seminar (yes, we know, we broke a pattern here: we should have named it “One World FLAT Seminars”, but this name is funnier).

The main goal of this project is to keep the community working in our area alive and updated, by bringing together researchers from all over the world in a virtual, accessible, and inclusive environment. We believe that recently our community is quite fragmented, so having a common platform to share old and recent results on the one hand would help established researchers working on similar topics to find collaborators and fresh ideas and, on the other hand, young people, new in the area, would have a clear vision of what is going on in this branch of theoretical computer science.

The talks will be accessible via Zoom and will run, at least at the beginning, on a monthly basis.

Speakers Wanted!

If you would like to volunteer to be the speaker at one of our sessions in 2027, please feel free to drop us an email with the details of your talk.

Upcoming talks

  • Formal Language Theory and Transformers

    Speaker: Michaël Cadilhac
    Post thumbnail
    Post thumbnail
    At the heart of the recent boom in AI lies a mathematical model known as transformers. They consist of many large matrices and nonlinear transformations, arranged in a fixed way, whose internal values and parameters are tweaked during the learning process. This leads to tantalizing natural questions: if I have a target formal language in mind, can these parameters be tweaked in any way in order to recognize this language? Is there a correlation between the number of parameters and the expressive power of the transformer? Does alternating more or less between linear and nonlinear transformations change the expressiveness? [Read More]