Hello - this looks like an awesome project, thank you for doing this.
I am interested in machine learning over graphs - because I think that many real world entities are well represented as a graph. I've been working with deepmind's graphnetworks library but I have got the idea that their approach is not really exploiting a graphical representation. I came across this package when looking for other takes.
However, I am somewhat lost! Could you recommend a survey or tutorial paper on the general approach so that I can prime myself before approaching the package and trying to get to grips?
Many thanks in advance for any pointer that you have!
Hello - this looks like an awesome project, thank you for doing this. I am interested in machine learning over graphs - because I think that many real world entities are well represented as a graph. I've been working with deepmind's graphnetworks library but I have got the idea that their approach is not really exploiting a graphical representation. I came across this package when looking for other takes. However, I am somewhat lost! Could you recommend a survey or tutorial paper on the general approach so that I can prime myself before approaching the package and trying to get to grips? Many thanks in advance for any pointer that you have!