cheekyfibonacci

What are everyone's thoughts on this paper? Does anyone understand category / topos theory willing to explain?

the authors seemed to have lost an opportunity to name the paper "In the Shallow-lalalow"

don't confuse this with universal approximation - yes shallow ReLU networks are dense in functional space, so at the limit you should be able to get any function you want - but they are talking about exact representation with finitely many neurons here.