[FoRK] Whole Brain Emulation: A Roadmap
jbone at place.org
Tue Jun 22 04:37:00 PDT 2010
Excellent overview by Bostrom. Should be required reading for further
discussion of the related topics, for folks who haven't already and /
or are not close to the subject matter.
I quibble with a few of the estimates etc., but think this is a fairly
conservative and useful overview. One example quibble: if you can
get away with only simulating certain structures at high fidelity,
such as the neocortex, and merely provide functional analogues of
others, the picture changes somewhat. And there is a big difference
depending on the resolution of the simulation of the relevant parts.
If you can abstract certain things like electrochemical potentials and
gradients, you get there cheaper and faster than if you have to
simulate down to the molecular level, dynamics and all.
More information about the FoRK