Re:Growing Neural Cellular Automata: A Differentiable Model ...
by openfuture on Tuesday February 11, 2020 @01:34PM
(#22301143)
Why?
[ Parent
| on HN ]
Re:Growing Neural Cellular Automata: A Differentiable Model ...
by m_que on Tuesday February 11, 2020 @03:57PM
(#22302764)
This is the foundation for 'programming' morphogenesis of multi-cellular synthetic life in the far future. Of course you first have to get the programming of monocelluar life to work.
Also, the reverse might be possible, i.e., decompiling genetic code from phenotype and so on.
Re:Growing Neural Cellular Automata: A Differentiable Model ...
by imvetri on Tuesday February 11, 2020 @11:19PM
(#22305737)
Foundation for programming the morphogenesis. Yes.
morphogenesis - concept is fundamentally flawed. It's not thoroughly put in theories. Computing has touring test. Does concepts like morphogenesis has any tests? Nope. So building something on top of half baked concepts will not scale/sustain/be a foundation.
If we really had understood morphogenesis in the past and the it is 100% put in theories, we would be having a virtual world mimicking biological world by now.
Re:Growing Neural Cellular Automata: A Differentiable Model ...
by m_que on Wednesday February 12, 2020 @02:03AM
(#22306557)
Yes, morphological models can be tested. You can hypothesize to reactions to alterations of an organism structurally and chemically (morphogenes).
Diffusion-reaction models are available for certain kind of morphogenetic behavior. See also https://www.brandeis.edu/now/2014/march/turingpnas.html
Re:Growing Neural Cellular Automata: A Differentiable Model ...
by dropit_sphere on Tuesday February 11, 2020 @06:25PM
(#22304081)
>Also, the reverse might be possible, i.e., decompiling genetic code from phenotype and so on.
I don't know anything about decompiling or ML, but doesn't the black box nature of NN rules this out?
Re:Growing Neural Cellular Automata: A Differentiable Model ...
by m_que on Wednesday February 12, 2020 @01:56AM
(#22306531)
NNs are not a black box in the sense that you cannot see what is going on in them, but in the sense that we currently just very badly understand. You have a function approximator that you have to translate into some other representation. As long as you know how basic building blocks map or can be approximated, it might be possible.