Scientific minds have all the time tried to determine methods to deliver mind’s processing capabilities to computer systems. Creating algorithms that may simulate the human mind and the way neurons work are essential to attaining that dream.
One such try is made by a staff of researchers primarily based in Germany, Japan, Norway, and Sweden. They’ve created a brand new algorithm that’s designed to simulate mind’s 100 billion neuron interconnections, i.e., nearly 100%. However there is no such thing as a laptop or supercomputer able to doing it.
The algorithm is created with the assistance of the open supply simulation device referred to as NEST (Neural Simulation Device) that’s broadly utilized by the neuroscientific group. Additionally, it’s the core simulator of the European Human Mind Mission.
The human mind simulation requires the digital neurons to be linked to nodes (tiny computer systems with a number of processors that do the calculation half).
Even with huge supercomputers, it’s subsequent to inconceivable to simulate 100% of the mind. With an earlier model of the algorithm, the researchers had been capable of reproduce just one% or 1 billion connections on their petascale Ok supercomputer.
The reason being that the reminiscence required per processor to simulate simply 1% of the human mind could be very excessive. If your complete mind comes within the image, the reminiscence necessities bounce to nearly 100 occasions per processor the present supercomputers have.
Sooner or later, with exascale supercomputers (having extra processors per node), it could be doable to scale the NEST algorithm to attain sooner whole-brain simulation. Then additionally, the reminiscence per processor and the variety of nodes would keep the identical. However the superior NEST algorithm would be capable of optimize the reminiscence required by the system.
The researchers have described their mind simulation algorithm in a white paper revealed in Frontiers of Neuroinformatics.