[whimsicalism]

adversarial noise is very popular in the media but imo is a complete dead end for the desired goals - representations do not transfer between different models this easily

[dijksterhuis]

adversarial noise [transferability] for image classification used to be very easy (no idea now, not been in the space for half a decade).

the [transferability] rates just drop off significantly for audio (always felt it was a similar vibe to RNN ‘vanishing gradients’)

edit — specifically mention transferability