Thread by @mmbronstein

In summer 2020, I started writing a blog on #GraphML in @TDataScience

towardsdatascience.com/graph-deep-lea…

It was a new and rewarding experience, from which I learned a lot. I was surprised that such a technical topic would attract >200K views and >6K claps in less than half a year.

I am grateful to all the readers and colleagues whom I used as "guinea pigs" to proof-read my posts especially @frederickmonti @emaros96 @b_p_chamberlain @ffabffrasca @gbouritsas @__lucab @fhuszar and to @kmborgwardt @thomaskipf @guennemann whose comments helped improve them

The first post was based on a seminar I gave for the @IEEEsps on the successes and challenges of #GraphML and #geometricdeeplearning

towardsdatascience.com/deep-learning-…

Then, a series of three posts on the expressive power of #graphneuralnetwork and the Weisfeiler-Lehman test

towardsdatascience.com/expressive-pow…

structural encoding

towardsdatascience.com/beyond-weisfei…

and isometric embeddings

towardsdatascience.com/beyond-weisfei…

A controversial post on whether depth helps #graphneuralnetworks turned out to be one of the most popular

towardsdatascience.com/do-we-need-dee…

and the post on deriving convolution from first principles of symmetry got most feedback

towardsdatascience.com/deriving-convo…

On the other hand, the post on the applications of #GraphML for the design of anti-#cancer #hyperfoods written with Kirill Veselkov and Gabriella Sbordone that took me the most time to produce did not attract as many views as I expected:

towardsdatascience.com/hyperfoods-958…

At the same time, a more recent post I published on new year's eve on the use of #GraphML in biological sciences was a hit

towardsdatascience.com/geometric-ml-b…

My resolution for 2021 is to continue writing on a broader range of topics as well as trying new formats (stay tuned). As usual, any feedback, especially critical is very welcome.