Slashdot Mirror
The Future of Optical Fibre
An anonymous reader writes "An Australian researcher has come up with a novel way of developing optical fibres. Steven Manos, a researcher at the Optical Fibre Technology Centre in Sydney, Australia has developed a method of using genetic algorithims for discovering optimal designs of optical fibres. An article on his work had this to say "The problem with designing optical fibres is starting with a specific set of criteria and then coming up with a design to fit this. The computer program developed by Manos, which is run on supercomputers, does this by mimicking the process of evolution. The computer program combines two patterns to create a third fibre 'offspring', which Manos described as "similar but a bit different". This process is repeated thousands of times with the 10 designs best suited for the particular application chosen to 'breed' again." Another case of "When in doubt, use brute force"?"
I'd rather not think of the method as brute force. Ok, its not like a design from first principles, but its still way to search the parameter space without having to test all coimbinations of parameters
Does his for lack of better words breading of fiber networks, did I understand this right, take into account some immovable obstacles?
Or is this process used to design the cable itself?
A psychopath can't tell the difference between right and wrong. A sociopath knows the difference - he just doesn't care.
I have doubts of the .edu going down, but just incase here's the google cache:
http://64.233.167.104/search?q=cache:9vv96kkSrhsJ: www.oftc.usyd.edu.au/+&hl=en
But he's already started off with intelligent design!
..and now this? What other uses could genetic algorithms have?
I don't suffer from insanity, I enjoy every minute of it! --Longbottle
There are some interesting PDFs of papers co-written by Steven Manos available including these two:
I'm not going to pretend that I know exactly what's going on, but the first of those two is worth looking at if you have even a passing interest. The second looks to be a little more towards the "deep end".
'Thats they exact same thing a banana wrench monkey.'
I mean its like every tosser who works out that their specific problem can be described by a phase space comes to the startling conclusion that a genetic algorithm is a neato way to explore the said phase space? I think I should start submitting patent applications of the form "Method and algorithm for optimising , using a fitness function and genetic algorithms to find optima in a phase space".
This is a much better example of the application of Genetic Algorithms than the story that was on slashdot the other day (can't find a link, the one about Formula One racing).
in this case they have a very specific set of criteria.
it didn't however mention in the article how they're testing the designs (did it?)...
and are they actually manufacturing any of the designs that have come from thiss yet?
and if you see me strut, remind me of what left this outlaw torn...
how is this, a fine achivement of artificial inteligence labs, that is programs that have sex, a brute force approach?
i refuse to believe that i was created by brute force, besides, dad is a nice guy
America != World
.. would breeding be regarded as "brute forcing". :)
Norman Cook's Ode to Sl
...wireless!! ;)
Another case of "When in doubt, use brute force"?
Evolutionary search isn't "brute force", you id... At least not for meaningful definitions of 'brute force'
Brute force would be starting at one end of design space and evaluating each design in turn.
Belief is the currency of delusion.
..as they don't search the state space exhaustively. Going through all possible combinations of parameters would be brute force, but in this case, as the parameters are real-valued, this is even impossible (if ignoring the possibility of quantisation)
Evolutionary Algorithms provide informed search as they perform competition among the individuals (each representing one possible solution) in the population. Their performance is way above exhaustive search techniques (which _are_ brute force) but below classical search techniques. In this case, however, such classical techniques cannot be applied as the problem space is not well-defined.
Another case of "When in doubt, use brute force"?"
No, Taco, No.
From the 'brute force' entry in Wikipedia:
In computer science, Brute Force, sometimes called the Naive Method, is a term used to refer to the simplest, most intuitive, most spontaneous, and usually most inefficient methods of accomplishing a task.
This is exactly what a genetic algorhthem is not. If you have a million numbers brute force would be to go from the first to the last in order. Using a genetic algorhythem provides a shortcut though Design Space wherein you need to try far fewer combinations in order to come to a successful result.
C'mon Taco, of all people, you should know this!
Big apple, new Yorik, undig it, something's unrotting in Edenmark.
With all the weirdo animals the Australian continent has produced I guess this program will produce some highly interesting results. I cant wait for the announcement that a pattern resembling a Duck Billed Platypus is ideal for streaming Digital TV.
Do not try to read the dupe, thats impossible. Instead, only try to realize the truth
What truth?
There is no dupe
Genetic algoroithms are simply another form of optimization algorithm, just like Simulated annealing, Ant-Cology optimization, just to name a few. Each variety has its strengths and weaknesses for different search spaces and genetic algorithms have there place. These often have nature related names because nature is an excellent optimizer from which we draw inspiration.
If you want to talk brute force, try an exhaustive search of complex high-dimension, continous, real valued parameter space. Get back to me when you are done.
Repeat after me: There is no general solution to the global optimization problem.
2^5
shouldn't you all be working
Here is the research paper published by Manos on the topic.
fifteen jugglers, five believers
We must kill it before it develops language skills!!
-- Thou hast strayed far from the path of the Avatar.
You may be trolling, but I'll point it out anyway: evolution doesn't have an end point. It is a path - a path with many branches and dead-ends - and not a destination. And I'm not being all metaphysical and hippy-ish about it.
Comment removed based on user account deletion
The (hic) Master (hic) would not approve of this...
Somehow I don't think he's going to cover it with bread crumbs.
ThE MaStEr DoEs NoT aPpRoVe Of WiReLeSs, OnLy FiBrE
Those who complain about affect & effect on
I am... Torgo..; I .. polish the. .. fiber while the Master... is away...
There is no way... out of.. here, the fiber will.. go dark ... soon, there is no way... out of .. here...
etc, etc.
Seems yet again there is a better way to deliver data over fiber, which doesn't surprise me. Does anyone remember when Tyco Corp. (http://www.tyco.com/ - the same guys responsible for a nasty embezzlement scandle) used to string fiber across the oceans like mad. And then someone figured out that you could send 100x the amount of data across the same cables...
Genetic algorithms are computational shortcuts that are used to very quickly find minima in complex multiparameter functions.
Suppose you wanted to find the lowest value of f(x)=sin(x) where x is from 0-360. (OK we all know its at x=270 but hear me out) - you can do it a couple of ways:
1. calculate sin(x) for all 360 possible values of "x" or
2. calculate sin(x) for (say) 20 values of "x".
Statistics says approach 2 will give you a couple of promising results, for only 1/18th of the effort. Now "breed" another 20 from the 6 values of x for which sin(x) were lowest, say 190, 210, 212, 260, 278, 290. This "next generation" gives sin(x) values whiach are closer to zero. Take the best 6 again. After three generations you are *close* to finding the values for "x" that give you sin(x)=0.
So systematic examination takes 360 tries and the genetic shortcut takes 60 tries - about 17% of the computational effort.
Now imagine a function a bit more complex; some mad multivariate affair like the wave equation. Each variable becomes a "gene" in the above "breeding program". All the time we are looking for parents and offspring that *tend* towards the answer we are looking for. (We also chuck in some unrelated parents too, since inbreeding can be bad - a tip stolen from Monte Carlo techniques [which see]).
The computational savings from GA, GP and MC techniques are potentially huge (as in orders of magnitude) so long as you dont care that:
a) The answer is not 100% exact
b) Some alternative minima are missed
I wish at was Friday, but I dont want to wish my life away. So I wish it was last Friday.
What is the purpose of this? Specialized interconnects? As far as I know, there is already massive overcapacity for telecom fiber networks.
For exploring real-valued phase spaces, one solution is to combine a GA with a classical hill-climber. A hill-climber evaluates the local gradient (the partial derivatives of fitness with respect to the independent variables) and then makes a directed adjustment of the solution in the direction of better performance. Hillclimbers can reach optima in floating-point spaces very quickly, but tend to get stuck on local solutions.
GAs are great for jumping out of local optima to find new realms of the solution space, but don't converge as quickly on the neighborhood optima. So the combination of a GA with more classical optimzation can work well.
Two wrongs don't make a right, but three lefts do.
GA and GP is inherently random, just pray your population doesn't become stagnent. Or you will need to repeat your experiment, all the time thinking, WTF!!!, brute force would have been faster ;P
I dropped out of AI class after about 3 months because my teacher was a complete moron. It's too bad, I liked it but she made class and the topic about as painful and agonizing as trying to nail jello to your own ass. I didn't make it to genetic algorithms
"Many words are similar in English and French as both languages evolved along side each other "
Actually it has more to do with the Normans invading england in 1066 and dumping literally thousands of french words into the english language. English is an unusual language in that its actually a true amalgamation of a number of languages (anglosaxon , norman french and old norse with a few others thrown into the mix too). Generally (linguists don't flame me ok) it has germanic verbs & nouns + norman french nouns and scandinavian grammar (eg verb comes before object in sentence)
I would not be surprised if this is the way our own brain works when figuring out problems. Raw ideas form and mutate and are tested - sometimes we are conscious of it, many times not. It's the way species evolve - it only makes sense that the same logic is built within our own brain.
Of couse, I am also a big proponent of the idea that evolution gave humans the greatest gift of all - the ability to self-evolve ourselves.
The government's moral compass is controlled by GPS.
In times of crises, they alter it to suit their needs.
Yeah, people can apply GA's to any sort of problem. You know why ? Because GA's are soooo nice :-) *but*, you have to (a) design your fitness function and (b) design your representation and (c) design your recombination and mutation operators.
__That's__ what takes ages.
How can anyone make a claim like this? Just the fact that one can't think of any other algorithm doesn't mean no such algorithm exists. For many problems that can be solved by genetic algorithms, other (problem-specific) algorithms exists (or may exists) that are way more efficient. The nice thing about genetic algorithms is that it is a standard tool that often works, not that it is an exceptionally smart way of doing things.
This process is repeated thousands of times with the 10 designs best suited for the particular application chosen to 'breed' again." Another case of "When in doubt, use brute force"?
More like another case of computer science being fascinated by meat.
Remember when neural networks were the next big thing? Everyone was applying them to everything, whether or not it made sense to solve the problem that way. It's neural! Just like our brains! Our brains are smart, they will make our computers smart!
I'm sure genetic algorithms will eke out a useful place in the computer science toolkit, I just doubt it will be as broad as the current fashion of applying them to everything from optical fiber to race cars to compilers.
Never approach a vast undertaking with a half-vast plan.
just remember that if you're installing this fiber, it takes 4hrs a foot, and there's annoying clank-a-jank music playing the whole time.
Keep your packets off my GNU/Girlfriend!
Structured Wild Ass Guessing.
At least they are using a computer to do it.
I think you underestimate just how much I just dont care.
Manos... the hands of bitrate.
One of the coolest things I ever saw that used Genetic Algorithms was the Pixar lamp character they use before movies. The model was composed of a simulated mechanical system and they used GAs to teach it how to jump realisitically (and not fall over).
I've written my open GA for doing circuit optimizations and it works very well. The thing I love most is that they are so simple to write. There are things you can do to speed up convergence but the basic algorithm is very straight forward.
The difficult thing is how to score individual trials. I don't know how many times I've checked things after a overnight run and found that my results aren't what I expected. Pretty much everytime this comes down to how I've scored a trial. Just remember you get what you ask for.
For a circuit example, suppose I ask for a certain power comsumption and speed but I overstate the speed goal. Because I'm so far off the speed goal the power will largely be ignored. There are easy ways to tweak this but the point is...again...you get what you ask for.
One should not theorize before one has data. -Sherlock Holmes-
However:
America > World
They conquered england. They didn't exactly make much headway in wales until the next century and never made it far into scotland.
on the next generation of fiber so i can upgrade my christmas tree.
i got a roll of fibre on the back of my car! gbit whereever i go, anybody wants to have a ride or what?
Steven Manos... I guess the "fate" of fibre is in his "hands."
Diplomacy is the art of saying, "Nice doggie!" until you can find a rock.
There are a few posts here to the affect that "optimization is limited completely by the original machines." When I was doing GAs, we would select the top few performers, and:
1. "Breed" them with each other.
2. "Breed" them with totally random data.
No matter how well your select your original machines, there's practically always room for improvement (otherwise, why use a GA in the first place?) Unless you are REALLY good at selecting your first few machines, the random data really is powerfull. Case in point, after a few million itterations, a GA designed to play Othello randomly generated the top performer. It was completely different than any other machine but capable of consistantly beating (granted by a very small margin) both other machines and people.
This tagline brought to you by 1500 monkeys in just under 17 years.
Troll? Figures huh, i shouldn't have been baited by the troll parent.
I thought i was being 'informative'.
Well, at least it makes A.I. sound like fun...
Darwin would be well pleased!
..so maybe plain 'ol meat isn't so bad.
While you're demonstrating ignorance, there is a lot of very promising work going into applications of neural networks to control systems and the broader field of AI in general. The problem with neural networks is that you need large numbers of processors to do some of the more complicated nets in anything approaching real time. Your brain has several billion little processors massively interconnected.
Up until very recently with the advent of large scale FPGAs, this has not been practical.
..don't panic
The article vaguely implied that maybe flexibility and durability was the issue. It also mentioned puting a pattern of little holes in the fiber. Does anyone know what those are for?
-jim
maybe he meant it was greedy.
No. Brute force would be making a list of all possible designs, removing the ones which did not fit the requirements, and sorting by price. This method explores only a small subset of all possible designs - while it won't find the theoretical best possible design, it'll find one good enough, and it'll do it in a timespan shorter than the age of the universe.
I like mine round, long, thin, and to the curb.
-- Having a Creationist Museum is like having an Atheist place of worship
Remember, folks...
"Manos"... the Hands of Fate.