top | item 34186690

Cell intelligence in physiological and morphological spaces [video]

55 points| hyporthogon | 3 years ago |youtube.com

13 comments

order

polishdude20|3 years ago

I've seen many of Michael's lectures and now I'm at the point of just waiting for something new from him. I'm super excited about his research but it would be nice to know if a new lecture has new material and discoveries in it without having to watch the whole thing.

carapace|3 years ago

Levin is undoubtedly doing some of the most important work in all of science, in my opinion.

Just as an teaser, one conceivable result of this work could be to unlock the secrets of regeneration! Imagine being able to regrow limbs and organs...

doitLP|3 years ago

I had the same thought. A great sci fi book that explores regeneration as an element is Pandoras Star (and it’s sequel) by Peter F Hamilton.

abudabi123|3 years ago

Related works are Lewis Wolpert's Triump of the Embryo, and the Dutch Flag problem.

agumonkey|3 years ago

I think he's handwaving too much stuff [0] but the experiments and path he (and the few in this niche) is pursuing .. seem to be opening a big crack in the biology map.

It's also ethically and existentially challenging because some of his results are frankly too wild, unless you planned to grow your own siamese head.

There's also a trend I'm sensing, that computing is more and more about tissue configuration (even outside biology).

In any case, considering the effect of better understanding of regeneration and cell homeostatis, I wish him and his team all the best.

[0] the API metaphor and fractal intelligence seems like magic terminology but I could never find any interview where he gave more details.

xoxo2212|3 years ago

> There's also a trend I'm sensing, that computing is more and more about tissue configuration (even outside biology).

I'm curious what do you mean?

Examples?

naturedivine|3 years ago

When we "control" computers from outside, there is nothing observable from within the computer that can denote the control.

In exactly same way, the control is happening from outside cells to make them achieve an objective, which also is set from outside. DNA are only like subroutines to create stuff in localised environment, but not orchestrators of how the overall system is evolving.

Thie type of source of control is in form of information, and that information flow must be top-down, without any cellular or other boundaries.

trebligdivad|3 years ago

He almost seems to be describing something 'declarative' - i.e. saying where things should endup/be arranged rather than the process of how to get to it.

chaps|3 years ago

21 minutes into this video and this is already one of the most interesting things I've watched in a very long time. Thanks for sharing.

dan_mctree|3 years ago

This is absolutely stunning. I'm beyond curious what is possible in this space. Could we potentially start regenerating and even redesigning lifeforms, including humans? How about inducing growth of a second brain? Or perhaps we could learn how to merge multiple already existing brains together and see if we can create one more effective brain

akomtu|3 years ago

Looks like he has discovered the "electric body" that tells cells what lifeform to create. It's an impressive milestone in science, but I hope he grows his cute xenobots in a bulletproof box. He should try to grow xenobots in very different environments and locations.

ljlolel|3 years ago

He had a great talk at NeurIPS and it’s amazing research