Later On

A blog written for those whose interests more or less match mine.

At last! I have found it again!

leave a comment »

I read this article some years back. It describes an experiment in using genetic algorithms on hardware, as it were. It worked great, except the solution was unique to the particular lab and hardware and local situation: the genetic algorithm exploited idiosyncrasies of the hardware at hand to find solutions that could not easily be replicated. Fascinating article. It begins:

Let Darwinism loose in an electronics lab and just watch what it creates. A lean, mean machine that nobody understands.  Clive Davidson reports.

“GO!” barks the researcher into the microphone. The oscilloscope in front of him displays a steady green line across the top of its screen. “Stop!” he says and the line immediately drops to the bottom.

Between the microphone and the oscilloscope is an electronic circuit that discriminates between the two words. It puts out 5 volts when it hears “go” and cuts off the signal when it hears “stop”.

It is unremarkable that a microprocessor can perform such a task—except in this case. Even though the circuit consists of only a small number of basic components, the researcher, Adrian Thompson, does not know how it works. He can’t ask the designer because there wasn’t one. Instead, the circuit evolved from a “primordial soup” of silicon components guided by the principles of genetic variation and survival of the fittest.

Thompson’s work is not aimless tinkering. His brand of evolution managed to construct a working circuit with fewer than one-tenth of the components that a human designer would have used. His experiments—which began four years ago and earned him his PhD—are already making waves. Chip manufacturers, robot makers and satellite builders are interested because the technique could produce smaller, more efficient devices than those designed today using traditional methods. Thompson’s experiments have also inspired other research projects and some serious speculation about whether technology is poised to evolve in ways that will take it well beyond human understanding.

Looking for inspiration

Computer scientists have long looked to biology for inspiration. From simplified models of the brain they developed neural networks that have proved particularly good at recognising patterns such as signatures on credit cards and fingerprints. They have also worked out ways to mate and mutate programs and allow the resulting programs to compete with one another to generate the “fittest” software for a task.  These “genetic algorithms” have been used to evolve software that does everything from creating works of art to selecting high-performing shares on the stock market.

To Thompson, who works with Phil Husbands at the Centre for Computational Neuroscience and Robotics at the University of Sussex, all these techniques leave something to be desired. They are too tightly constrained by the rules of chip designers and software engineers. The behaviour of living neurons, for example, is inseparable from the biochemicals from which they are made. But it doesn’t matter what material the circuits of a neural network chip are etched in, so long as they operate in a digital fashion. …

Continue reading.

Written by LeisureGuy

8 December 2008 at 2:11 pm

Posted in Technology

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.