Em progress: "40 Trillion Pixels…1000s of ultra-thin 40-nanometer slices … aligned … into a coherent, 3D image volume of the entire fly brain … not yet a true connectome since establishing a connectome requires the identification of synapses"https://ai.googleblog.com/2019/08/an-interactive-automated-3d.html …
-
Show this thread
-
Replying to @robinhanson
It is a very neat step forward, even though the real bottleneck will be identifying functional properties. It seems to fit my mental model that once big datasets available CompVision/ML people will invent decoding methods fast.
1 reply 0 retweets 11 likes -
Replying to @anderssandberg
The em-scenario hope is that we can identify cell types, to match to sufficient cell-type specific signal I/O models, without understanding that much about higher level structures and functions.
2 replies 0 retweets 5 likes -
Replying to @robinhanson
Yes, that is my view too. I sometimes call it the "IKEA assembly approach to software intelligence" - you do not know what all those parts are supposed to do, but if you follow the assembly instructions correctly it turns into furniture without any knowledge of carpentry.
4 replies 3 retweets 14 likes -
Replying to @anderssandberg @robinhanson
What worries me is how to do the initial bridging from scans to parameters in computational models: we can have living ground truth but it is hard (and unscalable?) to measure parameters, and searching parameter space to fit behavior looks infeasible except for supersimple cases
1 reply 0 retweets 1 like -
Replying to @anderssandberg @robinhanson
Adding molecular info to the scanning should help bridge the gap to physiology somewhat: https://arxiv.org/abs/1404.5103
2 replies 0 retweets 4 likes -
This is much more feasible in optical approaches, but not entirely inconceivable even in electron microscope approaches to scanning: https://arxiv.org/abs/1806.00075
1 reply 0 retweets 2 likes
This paper is also cool: https://arxiv.org/abs/1806.04793
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.