1) Your article’s main point is apparently unreasonable, as far as evidence goes, because Deep Learning is demonstrably getting more and more general, as we apply more and more biological constraints.

b) Reference A ~ “Towards deep learning with segregated dendrites”: https://elifesciences.org/articles/22901

c) Reference B ~ “Deep Image Priors”: https://www.youtube.com/watch?v=_BPJFFkxSbw

▪

▪

2) Deep Learning has evolved, is still evolving, and will likely continue to evolve. Deep Learning/ML shouldn’t be observed as some fixed unchanging paradigm.