💾 Archived View for gmi.noulin.net › mobileNews › 3297.gmi captured on 2023-01-29 at 06:51:51. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2021-12-03)
-=-=-=-=-=-=-
2011-08-19 04:49:21
IBM has developed a microprocessor which it claims comes closer than ever to
replicating the human brain.
The system is capable of "rewiring" its connections as it encounters new
information, similar to the way biological synapses work.
Researchers believe that that by replicating that feature, the technology could
start to learn.
Cognitive computers may eventually be used for understanding human behaviour as
well as environmental monitoring.
Dharmendra Modha, IBM's project leader, explained that they were trying to
recreate aspects of the mind such as emotion, perception, sensation and
cognition by "reverse engineering the brain."
The SyNAPSE system uses two prototype "neurosynaptic computing chips". Both
have 256 computational cores, which the scientists described as the electronic
equivalent of neurons.
One chip has 262,144 programmable synapses, while the other contains 65,536
learning synapses.
Man machine
In humans and animals, synaptic connections between brain cells physically
connect themselves depending on our experience of the world. The process of
learning is essentially the forming and strengthening of connections.
A machine cannot solder and de-solder its electrical tracks. However, it can
simulate such a system by "turning up the volume" on important input signals,
and paying less attention to others.
IBM has not released exact details of how its SyNAPSE processor works, but Dr
Richard Cooper, a reader in cognitive science at Birkbeck, University of London
said that it likely replicated physical connections using a "virtual machine".
Instead of stronger and weaker links, such a system would simply remember how
much "attention" to pay to each signal and alter that depending on new
experiences.
Artist impression of neurons connecting IBM's processor replicates the synaptic
connections between neurons found in the brain
"Part of the trick is the learning algorithm - how should you turn those
volumes up and down," said Dr Cooper.
"There's a a whole bunch of tasks that can be done just with a relatively
simple system like that such as associative memory. When we see a cat we might
think of a mouse."
Some future-gazers in the cognitive computing world have speculated that the
technology will reach a tipping point where machine consciousness is possible.
However, Dr Mark Bishop, professor of cognitive computing at Goldsmiths, was
more cautious.
"[I] understand cognition to be something over and above a process simulated by
the execution of mere computations, [and] see such claims as verging on the
magical," he said.
IBM's work on the SyNAPSE project continues and the company, along with its
academic partners, has just been awarded $21m ( 12.7m) by the US Defense
Advanced Research Projects Agency (DARPA).