I feel like you both (?) or at least just D thinks that this the first time I have come across this argument. It's a
I guess 1 way to see what we think: Do you consider what they do here rev eng or sci? Or both? https://arxiv.org/abs/1610.01644v1 …
-
-
will have a look soon and get back to you!
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I am not familiar with the field, and I only skimmed it, but the subject appears to be a branch of Graph Theory.
-
OK well, regardless of what you think it is, it's addressing/attempting to address the issue in the nature article.
-
& it's what I think also is a relatively useful way of examining scientifically the high-level/theory-level of (D)NNs.
-
So much work needs to be done to understand and not use brute force what these complex systems actually do.
-
And my point previously = scientific inquiry and not rev eng (since we know the low-level anyway) is what's required.
-
And this general topic of how and why we need to understand (D)NNs is what my PhD was on.
-
Also that day I had a migraine so it's likely that was affecting the comms as chronic pain does bad things to me.
-
I'll shut up on this issue for good though as I'm aware you're likely not as interested/invested as I.
- 11 more replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.