4 ms·

Please don't refer to them as black boxes. The internals are fully observable.

by etiam 9y ago

fragsworth 9y ago
It's a conceptual black box. There's no way for us to understand what each individual neuron is doing.
- etiam 9y ago
  Indeed? https://arxiv.org/pdf/1312.6034.pdf https://arxiv.org/pdf/1312.6034.pdf
- blt 9y ago
  The tools we have developed so far are limited, but that's different from "there's no way". Many academics are working hard right now to better understand deep neural networks.
ionforce 9y ago
Is there meaningful information in what one observes?
- aeleos 9y ago
  Yes, it turns out you can find meaningful information. etiam provided this https://arxiv.org/pdf/1312.6034.pdf https://arxiv.org/pdf/1312.6034.pdf The main issue is making sure what you are looking for is actually what the network is doing. You have to correctly interpret and visualize a jumble of numbers, which usually requires a hypothesis about how it worked in the first place. But assuming both go well you can gain meaningful information.
  - fjsolwmv 9y ago
    Can I train an NN to visualize the numbers?
andbberger 9y ago
Fully observable and we still have no idea what the hell it's doing. Makes neuroscience seem kinda bleak doesn't it? There has been a lot of great work lately building up a theory of how these things work, but it is very much still in the early stage. Jascha Sohl-Dickstein in particular has been doing some great work on this. We don't even have answers to the most basic questions. For instance (pedagogically), how the hell is it possible to train these things at all? They have ridiculously non-convex loss landscapes and we optimize in the dumbest conceivable way, first-order stochastic gradient descent. This should not work. But it does, all too often. Not a great example because there are easy hand wavy arguments as to why it should work, but as far as proofs go... The hand wavy argument goes as follows: - we're in like a 10000 dimensional space, for the stationary point we're at to be a true local minima that means each one of those 10000 dimensions goes uphill in either direction. It's overwhelming likely that there's at least one way out - there are many many different ways to set the params of the net for each function. Permutation is a simple example. We really have no idea how these things work. Anyone who tells you otherwise is lying to you...
- otabdeveloper1 9y ago
  > Fully observable and we still have no idea what the hell it's doing. Of course we do. It's matching a smooth multi-dimensional curve to sample data.

Please don't refer to them as black boxes. The internals are fully observable.

by etiam 9y ago

Please don't refer to them as black boxes. The internals are fully observable.

fragsworth 9y ago

It's a conceptual black box. There's no way for us to understand what each individual neuron is doing.

etiam 9y ago
Indeed? https://arxiv.org/pdf/1312.6034.pdf https://arxiv.org/pdf/1312.6034.pdf
blt 9y ago
The tools we have developed so far are limited, but that's different from "there's no way". Many academics are working hard right now to better understand deep neural networks.

ionforce 9y ago

Is there meaningful information in what one observes?

aeleos 9y ago
Yes, it turns out you can find meaningful information. etiam provided this https://arxiv.org/pdf/1312.6034.pdf https://arxiv.org/pdf/1312.6034.pdf The main issue is making sure what you are looking for is actually what the network is doing. You have to correctly interpret and visualize a jumble of numbers, which usually requires a hypothesis about how it worked in the first place. But assuming both go well you can gain meaningful information.
- fjsolwmv 9y ago
  Can I train an NN to visualize the numbers?

andbberger 9y ago

Fully observable and we still have no idea what the hell it's doing. Makes neuroscience seem kinda bleak doesn't it? There has been a lot of great work lately building up a theory of how these things work, but it is very much still in the early stage. Jascha Sohl-Dickstein in particular has been doing some great work on this. We don't even have answers to the most basic questions. For instance (pedagogically), how the hell is it possible to train these things at all? They have ridiculously non-convex loss landscapes and we optimize in the dumbest conceivable way, first-order stochastic gradient descent. This should not work. But it does, all too often. Not a great example because there are easy hand wavy arguments as to why it should work, but as far as proofs go... The hand wavy argument goes as follows: - we're in like a 10000 dimensional space, for the stationary point we're at to be a true local minima that means each one of those 10000 dimensions goes uphill in either direction. It's overwhelming likely that there's at least one way out - there are many many different ways to set the params of the net for each function. Permutation is a simple example. We really have no idea how these things work. Anyone who tells you otherwise is lying to you...

otabdeveloper1 9y ago
> Fully observable and we still have no idea what the hell it's doing. Of course we do. It's matching a smooth multi-dimensional curve to sample data.