On the other hand, once someone cracks that, a huge number of low-end jobs will be automated.
On the other hand, once someone cracks that, a huge number of low-end jobs will be automated.
A few weeks ago there was a paper posted on "Synthetic Gradients" which should make it much more practical to train RNNs for games. Before it required saving every single computation the computer makes to memory, which uses a huge amount of memory and computation. Using synthetic gradients they need only store a few steps in the past. And it can learn online.
And have some problems with PacMan as the system can't plan: https://www.technologyreview.com/s/535446/googles-ai-masters...
Maybe those should be the other way around?
Deep learning is still mere perception. It doesn't handle memory or processing, it just transforms input into output, typically trained by Big Data, way bigger than necessary statistically speaking, given the world we live in.
AGI requires super aggressive unsupervised learning in recurrent networks, likely with specialized subsystems for episodic and procedural memory, as well as systems that condense knowledge down to layers of the network that are closer to the inputs. At a minimum. And nobody is really working on any of that yet (or at least succeeding) because it's really damn hard.
That's why everyone in "AI" is rebranding as a deep learning expert, even though deep learning is really just 1980s algos on 2016 hardware - you gotta sex up feed forward backprop or you don't get paid.
Edit: to be fair, robot control is much simpler than AGI, and might be mostly solved with deep learning somewhat soon, I forgot the context of your post.
There's more going on than convolutional neural nets. Architectures with memory and attention mechanisms do exist.
As examples: https://arxiv.org/abs/1506.03134 https://deepmind.com/blog/differentiable-neural-computers/
What really doesn't exist is any meaningful stab at unsupervised (or self-supervised) training on completely unstructured inputs or any sort of knowledge condensation/compression, at least for time dependent problems. These are of paramount importance to the way we think, and to what we can do.
There's a lot of trivially low hanging fruit, too - I still have yet to see even a grad school thesis that starts with an N+M node recurrent network and trains an N node subnetwork to match the outputs based on fuzzed ins, and then backs that out into an unsupervised learning rule that's applicable to multiple problems. Or better, a layered network that is recurrent but striated, that tries to push weights towards the lower layers while reproducing the same outputs (hell, even with a FF network this would be an interesting problem to solve if it was unsupervised). These are straightforward problems that would open up new avenues of research if good methods were found, but are mostly unexplored right now.
I could be wrong, if I had real confidence that we were close I'd be working on this stuff, but I'm collecting a paycheck doing web dev instead...
Differentiable neural computers - https://deepmind.com/blog/differentiable-neural-computers/
> we introduce a form of memory-augmented neural network called a differentiable neural computer, and show that it can learn to use its memory to answer questions about complex, structured data
So it seems that deep neural nets can have memory mechanisms and be trained to solve symbolic operations.