I will do some digging for you tomorrow, it's 3:30 am and I really should get some sleep.
> Sure we know the first few layers do stuff like edge detection.
Yes, basic primitives.
> But the higher layers are much harder.
Well, those higher layers are built up out of the primitives the lower layers contain. So you get things that are conceptually easier to understand rather than harder.
> And even if you know images are activating them, writing code that can accomplish the same task is much harder.
That I completely agree with, I've basically gone that route in the last couple of months.
> The best NNs have already surpassed human accuracy at imagenet.
Yep. It's extremely impressive.
It's a watershed moment in the history of computing and half of us are still asleep or in denial. Which is fine by me. But please don't hype. It isn't magic and it isn't perfect - yet.
I will do some digging for you tomorrow, it's 3:30 am and I really should get some sleep.
> Sure we know the first few layers do stuff like edge detection.
Yes, basic primitives.
> But the higher layers are much harder.
Well, those higher layers are built up out of the primitives the lower layers contain. So you get things that are conceptually easier to understand rather than harder.
> And even if you know images are activating them, writing code that can accomplish the same task is much harder.
That I completely agree with, I've basically gone that route in the last couple of months.
> The best NNs have already surpassed human accuracy at imagenet.
Yep. It's extremely impressive.
It's a watershed moment in the history of computing and half of us are still asleep or in denial. Which is fine by me. But please don't hype. It isn't magic and it isn't perfect - yet.
edit: ok, found two of the things I've looked at:
https://stats.stackexchange.com/questions/114385/what-is-the...
https://stats.stackexchange.com/questions/146413/why-convolu...
And this is interesting code:
https://github.com/yosinski/deep-visualization-toolbox
If anything I think that NNs as applied to vision problems are somewhat easier to make sense of than doing the same for NLP or audio.
Though it definitely feels like cheating compared to actually programming a solution.
An alternative name for 'deep learning' would be 'automatic multi-level feature engineering', but that wouldn't be nearly as sexy.