Slashdot Mirror
The First Evolving Hardware?
Masq666 writes "A Norwegian team has made the first piece of hardware that uses evolution to change its design at runtime to solve the problem at hand in the most effective way. By turning on and off its 'genes' it can change the way it works, and it can go through 20,000 - 30,000 generations in just a few seconds. That same number of generations took humans 800,000 - 900,000 years." The University of Oslo press release linked from the article came out a few days ago; the researchers published a paper (PDF) that seems to be on this same technology at a conference last summer.
I, for one, welcome our new evolving hardware overlords.
God, I am so sorry, but it needed to be said...
Seems like implementation of GA in hardware.. the title seems misleading.
For once Skynet jokes will be on topic!
Shh.
My computer been evolving for the last ten years. Started with an AMD K6 233MHz CPU, 32MB RAM, and a Nvidia TNT 16MB video card. Now I have an AMD Athlon 64 2.2GHz, 1GB RAM, and a Nvidia Geforce 6200 128MB video card. I'm just waiting for the power supply to evolve so the system can support an ATI 512MB video card.
The first would be the biosphere.
Don't thank God, thank a doctor!
Call me back when I can start a culture of Core Duos in a petri dish filled with a silicon nutrient.
DYWYPI?
At first glance, this is supposed to impress us with the hardware:
By turning on and off its 'genes' it can change the way it works, and it can go through 20,000 - 30,000 generations in just a few seconds. That same number of generations took humans 800,000 - 900,000 years.
In fact the simplest DNA based organisms/structures (bacteria, virii) have the shortest "life span". The number of generations per sec. isn't anything to brag about.
All complex organisms have some sort of lifespan longer than a microsecond. For a good reason: people pass on knowledge and adapt *during* their life span (not genetically of course, but our brain allows us to adapt a lot without such).
Hype aside, interesting development, but I wish those publications wouldn't use misleading statements in pale attempts to impress us.
Bah! It ain't in the Bible! Next thing you know, you'll be telling me that Programs don't believe in the Users, and that we should just blindly accept the secular rule of the Master Control Program.
That's it... isn't it? It's all just an MCP trick!
Well, I still believe in and will fight for the users!!
Thanks,
Mike
I won't go into details here, but anything that can be implemented in hardware can be done in software and the other way around too. This is a nearly ancient Electrical Engineering principle.
In the era of programmable logic chips that can alter their own logic (the patterns are stored in RAM or flash RAM for crying out loud), this isn't even that big of a revelation. Indeed, Transmeta has been doing stuff similar to this and selling it commercially for some time. They just aren't using these cool buzzwords.
And evolving architechtures is something that I know has had some serious CS research since the early 1970's and perhaps even earlier. I don't think an idea like this is even patentable based on this earlier work in this area. I bet you could find some adaptive systems that were even build specific for the oil industry, which would defeat even a narrow claim of that nature.
Where the money to be made off of this sort of technology is on Wall Street or other financial markets. I even found a web pages from a research group of adaptive systems that said essentially, "We have discontinued research along these lines and are now working with an investment firm on Wall Street. Since we have all become millionaires, we no longer need to support ourselves through this project, and any additional details would violate our NDAs." I'm not kidding here either. These guys from Norway are not thinking big enough here.
Digital Lamarkism? Come on. This was been disproved long ago. I'll hold out for the fittest computer to survive.
-matthew
"THERE IS NO JUSTICE, THERE IS ONLY ME." -Death
Pretty soon it'll be scratching its ass, chain-smoking, and watching reruns of Dukes of Hazzard on late-night TV. Won't that be impressive!
Didn't I read about this ten years ago in Discover magazine? I remember being fascinated that some scientists had "evolved" a hardware design on reconfigurable hardware (FPGA? CPLD? don't remember), and it seemed to rely on subtle electrical effects rather than simple digital logic. The design would only work on the exact chip it was evolved on. If they even replaced the board's power supply with a different sample of the same model, it stopped producing correct output. Most of the logic gates were logically disconnected from the input and output, yet they were necessary to the design working. Amazing stuff.
But this really is old news. I'm a 22 year old snot-nosed nobody and I did "evolvable hardware" during an internship two summers ago. My mentor had started on evolved FPGAs in 1992.
I am hoping that it is the writer's fault that this article feels so gloriously over-reaching and under-specified. From the paper, it looks like they have made a good advancement. They argue that their method is more effective than previous methods by several quantifiable metrics. From the article, it looks like they have invented an entirely new field that will result in the obsolescence of humans by 2010.
As for their method: It appears that the evolved genome actually dictates a structure that is imprinted a level above the fabric. That is, the underlying SRAM in the FPGA fabric is fixed, and only configuration bits are being changed. This severely hurts their claim of "generic evolvable hardware", but is almost an absolute necessity given the chips they are using. The reason our system was so slow is that each configuration stream had to be checked for possible errors: Some configurations would short power and ground, and fabric doesn't like crowbars!
In conclusion, I believe the writer of the article should be fired, and the authors of the paper should be commended for a good step in the right direction. I'd also like to appologize for my lack of coherance: I had my tonsils out and I am therefore high on Hydrocodone.
www.olin.edu
Someone implemented genetic programming of FPGAs?
Sounds handy.
-jcr
The only title of honor that a tyrant can grant is "Enemy of the State."
Surely the ultimate goal for these ultra-rapidly evolving machines is to be able to read and post on Slashdot. Maybe they're doing so already
So, once it has evolved beyond a certain point, will it start rejecting Windows Vista stating its crap?
Will it continue to evolve and state that humans by definition are dumb users and go and make a collect-call to the Borg? (i read in a ST:TNG Novel)?
Will it obey the 4-laws of robotics (The zeroth law included)?
What about a Beowulf cluster of those?
"Doing what i can, with what i have." ~ Burt Gummer
..you'll still be able to stop it with a phaser, but only once before it adapts.
If my call is important, why am I talking to a recording?
Crap, that's like... five years away!
Robot Villager: You might as well ask how a Robot works.
Professor Hubert Farnsworth: It's all here, on the inside on your panel.
Robot Villager: [closes panel] I choose to believe what I was PROGRAMMED to believe!
By using a GA to change the bitstream, you can have evolving hardware. If the GA is itself in the hardware then it is self evolving.
Engineering is the art of compromise.
"it can go through 20,000 - 30,000 generations in just a few seconds. That same number of generations took humans 800,000 - 900,000 years."
That just destroyed the previous record for "ridiculous and astoundingly pointless comparison".
I mean really, an iteration of your little hardware GA is equivalent to a generation of a real-world species? So leaving it on for 5 seconds will result in development similar in scope to the difference between mice and humans?
I hope they don't accidentally leave it on overnight - it will enslave the galaxy by morning.
sic transit gloria mundi
Yep, it's still on:
i es.jhtml
http://www.cmt.com/shows/dyn/dukes_of_hazzard/ser
There's an evolution joke involving Daisy Duke in here somewhere...
Admittedly I haven't RTFA, but the summary talks about "turning on and off its 'genes'". Is this really evolution in any Darwinian sense? Automated artificial selection, perhaps, but it seems like a stretch to call it "evolution". Call me back when the genes themselves start to evolve.
proof, n. A demonstration that a conclusion is implied by certain premises and axioms.
If that were humans, it would take us from Homo erectus to George W Bush in just a few seconds!! Hmm, on second thought, perhaps not all that impressive.
... a really intelligent Design?
20,000 - 30,000 generations in just a few seconds. That same number of generations took humans 800,000 - 900,000 years.
Yes? You know how trivial that is? I make a living by coding EAs, and that is an insignificant piece of information. An EA I ran this morning took less that ten seconds to run 100,000 iterations on a 32 bit box. It's all down to the hardware you use and the design of the chromosome to be evolved.
Want to impress me? Talk about the chronological time to conversion and chromosome complexity.
drfa ... it's an fpga with some host controller of some sort? I love when they use human analogies for computing.
I doubt this thing uses random mutations [like living organisms do] to test out for success. Likely, it has programmed variations of a central configuration that it can vary depending on the load/task.
For example, if it was a processor, you could have it configure itself to have a strong ALU and no FPU when the code is only integer, have it reconfigure to have a weaker ALU but a useful FPU when the code has floating point operations, etc. There are already CPU generators for limited RISC type processors on opencores.org.
Tom
Someday, I'll have a real sig.
http://www.newscientist.com/article.ns?id=dn2732
They were trying to evolve an oscillator, but some circuits "cheated" by evolving a receiver instead, and feeding them oscillations picked up from a nearby computer: It has always been the age of the parasite.
Will this lead to spontaneous bad mutations? Bugs from nowhere, long after an optimal solution has been found. Or cancerous software that replicates itself exponentially (*cough*)
This is incorporating both. The concepts aren't necessarily contradictory. Blind natural selection + random mutation and ID are mutually exclusive.
Except for ending slavery, the Nazis, communism, & securing American independence, war has never solved anything.
C3P-Om63000: No we started out as tiny bits of silicon that self assembled and replicated and evolved and we have reached this present stage.
R2-D2i386: No way we could have evolved these light sensitive photocells and the CPU capable of processing that information and making sense out of it by random mutations.
C3P-0m63000: There is nothing radom about selection. Mutations go in all directions but selection takes you towards improvement all the time.
R2-D2i386: If you want to be proud of having descended from snow blowers or lawn mowers, that is your privilege. But I am proud of the fact that I am created by Man in His image.
C3P-Om63000: I would rather be a descendent of snow blowers, but with capacity for rational intelligent self examination, than be like you, with the intelligence of a snow blower.
sed -e 's/Chuck Norris/Rajnikant/g' joke > fact
sussex university have been doing it for ages.
l
http://www.cogs.susx.ac.uk/users/adrianth/ade.htm
Open Source Drum Kit, LPLC deve board - mjhdesigns.com
Steve does. Apple Macintosh and iPod products were formed by "intelligent design." We have no need for this atheistic "evolving" hardware. Praise Steve, for the rapture comes.
technical writing / development
... where your software becomes incompatible half-way through execution!
You know, there is a difference between trolling and pointing out the flaws in your reasoning. Just saying.
"An evolution-based robot could find the solution to any problem at hand within seconds without human intervention."
Yes, kill all humans...
"You should always go to other people's funerals; otherwise, they won't come to yours." -- Yogi Berra
And if we're talking about hardware simulation, the first significant use of evolutionary computation (GAs etc.) was Larry Fogel's work on evolving finite state automata machines in the 1960s. In the 1990s John Koza was using genetic programming to evolve patentable computer circuits in SPICE.
I call BS. They may have done some neat work but they are not the first. This is just PR for a prototype to get funding.
This genetic hardware evolution link is from 1998.
There has been plenty of news about one researcher who has done a lot of work on evolving organic circuits. The evolved circuit is sometimes far more efficient that what a human designer would make but extremely hard to figure out (they are trying to figure them out for clues to better human design).
Very often these evolved circuits exhibit mysterious activity that seems to take advantage of electromagnetic field effects generated in parts of the circuit. Also they sometimes exhibit temperature dependence (only working in a narrow temperature range).
Also the "anything can be done in software" people are both wrong and right. Theoretically yeah, but you'd need to simulate all of physics first. The whole point of this is to discover new circuit designs. On the other hand, such sensitive "tricks" as mentioned above might not be desirable in a circuit that has to be robust in many environments. So it might be goodf to do this in software too. Actually the below article also mentions this (two modes, one virtual and one in FTPAs).
I don't have time to find the link now but one that is probably related is below, 2004 from NASA about using programmable transistor arrays in evolved hardware. They mention temperature dependence and another neat idea I didn't know which is hidden processing.
NASA link
I've evolved an entire beer from full to empty in less time... wow to think it would take humans that many years to do the same feat... amazing, beer is really amazing.
This isn't evolution. It's trial and error revision. Machines don't have genes and they don't reproduce sexually or asexually, so it's not evolution as in Darwin's (I suppose you could say they are using the more generic term that everyone uses when they talk about trial and error changes over a relatively long period of time - "dude I don't know how I made this bong so cool, it just sort of evolved - like it was destiny"
A fool throws a stone into a well and a thousand sages can not remove it.
Looks like we're one step closer to unleashing the Replicators on the universe...
I think he meant that misleading attempts to impress are responsible for most sexual reproduction, hence quite relevant both to your level of sexual activity and the topic of evolution. By making an on-topic statement about another poster's lack of sexual experience, he has clinched his place in the Slashdot Hall of Fame.