IEEE Spectrum December, 2010 - 36

Hardware vs. Wetware

To understand the difference between the architecture of the brain and a standard
computer, compare the path of a hypothetical bit of data in a brain with that in a brain simulation.

Brain In the mammalian brain,
storage and computation happen
at the same time and in the same
place. Neuron 1 sends a signal down
the axon to Neuron 2. The synapse of
Neuron 2 evaluates the importance
of the information coming from
Neuron 1 by contrasting it with its
own previous state and the strength
of its connection to Neuron 1. Then,
these two pieces of information-
the information from Neuron 1 and
the state of Neuron 2's synapse-
flow toward the body of Neuron 2
over the dendrites. By the time that
information reaches the body of
Neuron 2, there is only a single value-
all computation has already taken
place during the information transfer.

Dendrite
Synapses

Axon

Computation

Neuron 1

Electrical
impulse

Neuron 2

computer On a computer, the

memory and processor are physically
separated-a significant physical
distance separates the areas where the
data is stored from the areas where it
is manipulated. Modeling just a single
synapse requires the following to happen
in the machinery: The synapse's state is in
a location in main memory. To change that
state, a signal must originate somewhere
on the processor, travel to the edge of
the processor, be packaged for transfer
over the main bus, travel between 2 and
10 centimeters to reach the physical
memory, and then be unpackaged to
actually access the desired memory
location. Multiplying that sequence by
up to 8000 synapses-as many as in a
single rat neuron-and then again by the
brain's billions of neurons yields a single
millisecond of brain activity.

Data
Bus
CPU

ing it with its own previous state and the strength of its connection
to Neuron 1. Then, these two pieces of information-the information
from Neuron 1 and the state of Neuron 2's synapse-flow toward the
body of Neuron 2 over the dendrites. And here is the important part:
By the time that information reaches the body of Neuron 2, there is
only a single value-all processing has already taken place during the
information transfer. There is never any need for the brain to take
information out of one neuron, spend time processing it, and then
return it to a different set of neurons. Instead, in the mammalian
brain, storage and processing happen at the same time and in the
same place.
That difference is the main reason the human brain can run on
the same power budget as a 20-watt lightbulb. But reproducing the
brain's functionality on even the most advanced supercomputers
would require a dedicated power plant. To be sure, locality isn't the
only difference. The brain has some brilliantly efficient components
that we just can't reproduce yet. Most crucially, brains can operate at
around 100 millivolts. Complementary metal-oxide-semiconductor
logic circuits, however, require a much higher voltage to function
properly (close to 1 volt), and the higher operating voltage means that
more power is expended in transmitting the signal over wires.
Now, replicating the structure we've described above is not totally
impossible with today's silicon technology. A true artificial intelligence could hypothetically run on conventional hardware, but it
34

NA * iEEE SpEctrum * DEcEmbEr 2010

Main memory

would be fantastically inefficient. Inefficient hardware won't stop us from running neuromorphic
algorithms (such as machine vision), but we would
need an entire massive cluster of high-performance
graphics processing units (GPUs) to handle the parallel computations, which would also come with the
power requirements of a midwestern college town.
So how do you build something that has an architecture like the brain's? Here's DARPA's gambit: Change
your architecture to merge memory and computation.
The memristor is the best technology out there for the
task. That's because the memristor is the first memory
technology with enough power efficiency and density
to rival biological computation. With these devices, we
are confident we can build an AI that can approximate
the size and power requirements of a mammal's brain.

Partly to avoid the folly of trying to coax
intelligence from fundamentally dumb hardware,
DARPA launched a program called SyNAPSE
(Systems of Neuromorphic Adaptive Plastic Scalable
Electronics) in 2008. The timing was good. That
year, HP Labs had created a functioning memristor,
a device hailed as the fourth fundamental electronic
spectrum.ieee.org

http://spectrum.ieee.org

Table of Contents for the Digital Edition of IEEE Spectrum December, 2010