anillacnn
We thank the reviewers for their attention and constructive feedback that will improve the quality of our work
We thank the reviewers for their attention and constructive feedback that will improve the quality of our work. We acknowledge the need for clearer referencing to the SM and will improve this. Novak et al. 2019 on the role of locality is indeed relevant; we will discuss it in the revised version. We agree that studying the effect of further training methods and architectures would be very interesting. One step we took in this direction, motivated by reviewers' suggestions, is to replicate the same experiment with Adam instead of fixed learning rate SGD.