“Computers can perform segmentation faster than the human eye, which cuts down the time it takes to trace neurons to a matter of minutes or hours. But they aren’t as accurate: algorithms can miss out bits of neuron or incorrectly merge two neurons into one. People are therefore still needed to check the reconstruction. Seung is tackling this requirement through crowdsourcing and, specifically, an online game called Eyewire, in which players are challenged to correct mistakes in the rough draft of a connectome. ”
> “The enormous amount of time, effort & money that goes into such projects might be overkill. I don’t need to know the precise details of the wiring of each cell and each synapse in the brain. What I need to know, instead, is the organizational principles that wire them together.”
I think this endeavor might still be useful, e.g., it may help identify missing elements in a model. An example of that in another field: astrophysicists did know the organizational principles to build galaxy kinematics models, but only using the Schwarzchild's method---which is akin to mapping the galaxy---they were able to confirm that supermassive black holes should be at the center of most galaxies. Hence, both approaches (machine learning and mapping the brain) may walk hand in hand, helping identify such organizational principles.
https://eyewire.org/explore
“Computers can perform segmentation faster than the human eye, which cuts down the time it takes to trace neurons to a matter of minutes or hours. But they aren’t as accurate: algorithms can miss out bits of neuron or incorrectly merge two neurons into one. People are therefore still needed to check the reconstruction. Seung is tackling this requirement through crowdsourcing and, specifically, an online game called Eyewire, in which players are challenged to correct mistakes in the rough draft of a connectome. ”