11–20 of 40 posts
Maybe this will help.
With articles like this, I want a "check back in 2 years" reminder, to see how the science shakes out. I'm not smart or informed enough to judge these current events style updates for myself.
With articles like this, I want a "check back in 2 years" reminder, to see how the science shakes out. I'm not smart or informed enough to judge these current events style updates for myself.
> Nonetheless, Hinton and a few others immediately took up the challenge of working on biologically plausible variations of backpropagation. Trying to prove the plausibility of a theory is one approach to science I guess... The researchers have already concluded that brains are simply information processing machines and that AI techniques are a sufficiently representative model to use to learn what brains are like. I…
You're downvoted but this is correct. It is much like when the analogy was "springs and cogs", and an academic department created in that era "cog-nitive" science, would be the attempt to rotate enough gears in the right way. Many presumptions are being made here in "computational cognitive science" which preclude including many relevant features of animal learning and animal biology. Their whole world view is that "…
Other things current AI's are lacking besides growth: embodiment + the social and physical environment, ability to make interventions in the environment, self reproduction, learning from reward signals, autonomy, adaptation, radical open-endedness.
"Patterns of electrical signals in neurons" are just part of the picture. Yes, learning happens there, but learning is fed by signals from the body and environment. It would be silly to focus on the neurons while ignoring the actual content, then start wondering where meaning comes from, and if syntax is enough. Meaning doesn't come from mere neurons, it comes from being an embodied agent.
1. Biological plausibility of back prop.
2. The lack of interest/consideration of time-continuous input on network. They are currently discrete and "learning" and inference is done separately. That's not how most organisms work.
3. The lack of consideration how brains (architecture, not weight) grows.
I might just be me missing something but I really have hard time seeing how things would scale in real world (ex: in Robotics applications of Neural nets) without those things addressed
There's three things I've always been baffled by the lack of interest in the current deep learning based AI field when it comes to parallels with biological brain: 1. Biological plausibility of back prop. 2. The lack of interest/consideration of time-continuous input on network. They are currently discrete and "learning" and inference is done separately. That's not how most organisms work. 3. The lack of consideratio…
Personally I don't believe chasing perfect biological plausibility will be very fruitful (in short term). An algorithm that runs efficiently on wetware will probably not be very efficient on current hardware like gpu's. The reason deep learning is so successful is for a large part that they are very good at exploiting the efficient linear algebra devices we have at our disposal (transformers are only the latest evidence of this).
> Nonetheless, Hinton and a few others immediately took up the challenge of working on biologically plausible variations of backpropagation. Trying to prove the plausibility of a theory is one approach to science I guess... The researchers have already concluded that brains are simply information processing machines and that AI techniques are a sufficiently representative model to use to learn what brains are like. I…
You're downvoted but this is correct. It is much like when the analogy was "springs and cogs", and an academic department created in that era "cog-nitive" science, would be the attempt to rotate enough gears in the right way. Many presumptions are being made here in "computational cognitive science" which preclude including many relevant features of animal learning and animal biology. Their whole world view is that "…
Actually, the mechanisms are chemical processes involving trophic factors (i.e. inputs to those processes) and alteration of the physical structures the signals are transmitted with. You say "the brain grows" but the alteration of its structure to strengthen our weaken transmission and connections in response to signals is how it grows usefully. Which was present in the work described by the article.
I may be missing something, but it’s just a click bait title with no substance.
The title seemed a bit click-bait-y to be too though.
The author almost makes this sound nefarious or short sighted. Workshops and symposia get rejected all the time for a mundane reason: Too many submissions for the available schedule resources at the conference. Important research gets "rejected" all the time, and the selection committees are not saying your topic/research are silly, illegitimate, or fantasy.