Detecting Groundwater Pollution Source Through Simulated Evolution

I have not yet seen an application of evolutionary computation like this one: Tracking groundwater pollution to its source. But it seems they also apply other kinds of soft computing (neural networks and simulated annealing).

Excerpt:

They point out that reliable and accurate estimation of unknown groundwater pollution sources remains a challenge because of the uncertainties involved and the lack of adequate observation data in most cases. The non-unique nature of the identification results is also an issue in finding the original source of a pollutant. They have tested the validity of different optimization algorithms including a genetic algorithm, an artificial neural network and simulated annealing and hybrid methods. All of these methods essentially process available data including pollutant concentrations and how these change over time and any monitoring data to home in on a potential source. The benefit of using such algorithms is that as more information becomes available another iteration will take investigators closer to the source.

It seems an interesting application of artificial intelligence.

New Book On The Block

Jason Brownlee has published a new book on intelligent algorithms and machine learning: Clever Algorithms Nature-Inspired Programming Recipes.

I read the topic on evolution strategies and genetic algorithms it is a well crafted text for quick reading, rapid implementation, code examples and good references at the end of each topic.

You can download the PDF, read it online or buy a real copy!

Professor Hans-Paul Schwefel Will Receive The Frank Rosenblatt Award 2011

From the press release:

"The 2011 IEEE Frank Rosenblatt Award will be presented at CEC 2011 (www.cec2011.org) to Dr. Hans-Paul Schwefel, Professor Emeritus, Chair of Algorithm Engineering, Faculty of Computer Science, Dortmund University of Technology, Dortmund, Germany for pioneering contributions to evolutionary computation through the theory and application of evolution strategies. Schwefel is a German computer scientist and professor emeritus at Dortmund University of Technology, where he held the chair of systems analysis from 1985 until 2006. He is one of the pioneers in evolutionary computation and one of the authors responsible for the evolution strategies. His work has helped to understand the dynamics of evolutionary algorithms and to put evolutionary computation on formal grounds.

Congratulations, Professor Hans-Paul Schwefel!

Old School Genetic Algorithms News On Google News Archive

Interesting feature this one from Google News. You can browse old newspapers and read old news.

Looking for genetic algorithms, Google News Archive gives you a nice amount of news dating back to 1980s.

For example, see this simulated annealing news on the excellent magazine Dr. Dobbs -- genetic algorithms and other optimization approaches are cited and/or briefly explained. It's from 1989.

This other one is a brief summary of students and researchers work on engineering mentioning genetic algorithms and the early work of David E. Goldberg during his good times at The Clearing House For Genetic Algorithms. The excerpt:

James Gleick wrote an article on artificial life during its early steps, citing genetic algorithms along the way.

And this another one suggests someone was, as early as 1983, already using genetic algorithms to do stock trading or something related.

Evolving Real Robot Swarms Through Simulated Evolution

Nice video this one: Deployment of Large Aerial Swarms.

It seems the authors applied an evolutionary algorithm (genetic algorithm?) to do the control of those robots.

Evolving Astronomy Data Mining Through Simulated Evolution

Interesting stuff this one: The Future of Astronomy is Automated.

It seems genetic algorithms may give another kind of role for traditional astronomers -- and not only for them, but also for other kinds of data miners.

Keep your eyes wide open data miners!

Evolving Car Racing Through Simulated Evolution

Nice pointer from Martin Pelikan's blog. There is an interesting video of the simulated car racing.

Genetic Algorithms, Complex Systems, Economics. . .

Interesting on-line paper this one. It shows how the Santa Fé Institute's researchers, genetic algorithms, and complex systems rediscovered a classical way of economics -- but, of course, there is more than just that on the paper. It's worth reading!

Have you ever wondered how free markets could be self-adaptable?

Genetic Algorithms, Proteins, Nuclear Power And More!!!

A nice summary of a new book on natural computing and, of course, genetic algorithms are a part of it: The Lessons Of Living Things.

Well, it is not a novelty, since computers have been helping us to build things and achieve results since the early beginnings of computing. Surely, the natural approach to computation was a nice move and helped to open a new and wide research field.

Rosenbrock Function And The Steepest Descent

It's interesting how the steepest descent method sweats a lot if not correctly set up inside the "right" parameters' interval of the Rosenbrock Function:

It's a well known testbed optimization function and evolutionary algorithms can find its optimum with no problems at all -- surely, as long as you try a traditional parameters' setting that any introductory Evolutionary Computation book gives you.

Using the steepest descent and setting it up as:

x = 2

y = 2

Step Size = 0.001

Stop Criterion = 10^-6

I got the following results (see the image below):

Pay attention to the path the steepest descent takes until finding the optimum at x = 1 and y = 1. It is said that an evolution strategy would follow a similar path if its population could be infinite. Since, so far, there are no real computers with infinite memory, such assumption cannot be verified in the real world.

During a second run I set the parameters x = 5 and y = 5. The optimization path run very far away from the previous one! And from the optimum too!

A Bright Future For Soft Computing

Interesting this article here: Rebirthing HAL.

I think I have been hearing the very same arguments since my first days -- and even earlier -- as a freshman at my former university.

Evolutionary algorithms, neural networks, and others bio-inspired approaches are just profound simplifications of real systems. Don't expect them, at least in the near future, to replicate what their real counterparts do.

Soviet Professor Joins The Timeline Of Evolutionary Computation

Breaking News

The Soviet evolutionary computation researcher and pioneer, Professor Trollov Buraninev, has been endowed into the Evolutionary Computation Timeline. His pioneer work on the theory of evolutionary algorithms was started much earlier than his Western nemesis' researchers on the same field. A list with some achievements by Professor Buraninev:

1. The Schema Theorem
2. The Building Block Hypothesis
3. The Theory Of Evolution Strategies
4. The First Evolutionary Programming Implementation
5. Genetic Programming

Of course, there are many others. The list above is just some of his breakthroughs.

Professor Buraninev states his work is much better and general than "those stuffs made by mainstream German and American researchers! I found a way to optimize any kind of function under any conditions without calculating the covariance matrix or any quasi-Newton-like approach!".

What about Bayesian approaches? Professor Buraninev is categoric: "Bayesian approaches are for n00bs! I like this Western fad of writing words using numbers rather than the letters themselves! I learned it with my would-be girlfriend! If were not for that Californian guy who fooled me, now I would be dating a checkers master girl! Concerning Bayesian approaches, well... In Soviet Russia you can know a posteriori probabilities even without knowing any a priori odds! And vice-versa!"

The director of the American part of the Antarctica special evolutionary computation laboratory, Professor Pangloss Van Thunder-ten-Tronckh, states: "Professor Buraninev's achievement is the direct result of the ice mapping of Antarctica we have been doing since early 1970s. We applied our very special evolutionary algorithm to perform that task. Below, there is the final map we got after some simulation years."

Professor Van Thunder-ten-Tronckh claims that "what you mainstream climate researchers have been argued as global warming is just the work of our team to take some ice from Antarctica and use it in our evening drink parties. The ice sheet shrinking of Antarctica is not more than our continuous 'work' at that continent. There is no global warming! I know this is an inconvenient truth..."

Professor Van Thunder-ten-Tronckh and Buraninev showed us some footage of their Antarctica research place. In the words of Professor Van Thunder-ten-Tronckh: "This footage proves that our laboratory is 100% real! We have been not involved in mainstream evolutionary computation journals because our funding came from CIA and the former KGB and these malevolent agencies forbade us of publishing our work! This picture below shows our supercomputer Metroidska Oblast 5000 proving that P = NP."

"On the faraway foreground, there is our private satellite for cable TV and other 'secret' stuffs, such as the Russian Brides Channel and the Poker Channel."

"This one is very interesting! Our secret satellite is sending us our daily work schedule. Proving that P = NP, making record complexity time algorithms, and answering the Big Question Of The Universe are common tasks we are used to do."

"Here, Professor Buraninev is looking for the aforementioned daily schedule. An uncommon sight, since it is usually done by our grad students and other slaves workers."

"This one shows Professor Buraninev just after having found our daily schedule."

"This one is extremely interesting! It probably shows some Western evolutionary computation researcher trying to take off from us our plans and techniques! I don't know why those guys wear boots that leave strange words printed on snow, such as deg511, HPS, jmerelo and so on!"

"These ones are our slaves grad students crew. They are enjoying their weekly five minutes spare time looking up to the sky -- it's a hobby of theirs! These guys are amazing!"

"The last ones show our main lab site. In this place so many evolutionary computation revolutions took place! It's a pity we have never been allowed to publish our findings... But one day we will be acknowledged as we deserve!"

After all these years, what are the plans of those two pioneers? Let them reply! Professor Pangloss Van Thunder-ten-Tronckh states: "I'm thinking of retiring... I'm tired of doing such a tremendous contributions to evolutionary computation without any sign of acknowledgement! I'll buy a beautiful house at Vladvostok Station  or at Nunavut to spend my Summer time there!"

The Soviet evolutionary computation genius, Professor Buraninev, has a different feeling: "The joint evolutionary computation mission made possible to me to have contact with the business friendly ideas from American society. Therefore, I'll invest my money on a start-up bussiness of Russian Brides! Any man will be able to find a Russian bride through evolutionary computation. I'll use all my knowledge to make possible to every man in this planet to marry a Sharapova-like woman!"

The Antarctic evolutionary computation lab is taking a day off to celebrate this date.

An Old School Genetic Algorithm News

Interesting finding this one! Just click on the picture above and read the news! It's from 1986 and published on Tuscaloosa News. I think it summarizes some aspects of the good times at The Clearing House for Genetic Algorithms. Very old school, indeed.

Genetic Argonaut Blog Has Got Its First Publication

That's it, fellow readers! This blog has got its first publication! :)

This feat owes a lot to our friend Pier Luca Lanzi. Pier Luca kindly invited me (and Professor Schwefel too) to publish the interview I made with Professor Schwefel when evolution strategies celebrated their 45th anniversary.

I am astonished as I had never thought the interview could be published in an official newsletter of such an excellent lineage such as ACM SIGevolution.

If you would like to read the newsletter, see it here. Pier Luca made, as usual, a beautiful work and if I were you I would not miss a single issue of SIGevolution!

Thank you so much, Pier Luca, Professor Schwefel, Professor Eurípedes, and all the SIGevolution guys and gals, for this amazing opportunity!

Genetic Argonaut Has Got Its First Citation

That's it, my fellow readers! This blog has got its first citation in an academic paper. See the citation here.

The paper is entitled "A comparison of selection, recombination, and mutation parameter importance over a set of fifteen optimization tasks". The paper is in the GECCO 2009 Proceedings

If you would like to cite this blog, feel free to do it!

Thank you so much for the authors who cited Genetic Argonaut!

A Nice Genetic Algorithm Lecture

Professor David Edward Goldberg (@deg511) delivered a nice lecture on genetic algorithms in which he shares his views on them.

Here you are the lecture in four parts:

Part 1

Part 2

Part 3

Part 4

I have not seen everything yet. As soon as I have done it, I'll share my views on the lecture.

Mega Man + Genetic Algorithm Vs Air Man

There is a video at YouTube in which the author claims to have developed the fighting style of Megaman using a genetic algorithm to win over Air Man (the boss of the sky-themed level in Mega Man 2).

It is interesting to notice that, over the generations, indeed, Mega Man deals more damage to Air Man. The fitness function is very simple:

F = MMLB - AMLB

MMLB = Mega Man Life Bar
AMLB = Air Man Life Bar

Negative values mean Mega Man failed to win. Otherwise, he was successful.

Just after 10 generations, Mega Man manages to win his first battle and in the next ones his skills are improved.

I wonder if Dr. Light used genetic algorithms to design its robots. If so, he would be a pioneer of evolvable hardware!

Evolving Bugless Programs Through Simulated Evolution

A brief and nice overview of a genetic programming application for debugging real computer code programs. See it here.

It's almost a surprise to me the programming language used was the good old C: A Real Programming Language Hero!

Ban Genetic Algorithms!!!

Credit: Arthur Gretton


During the last G-20 Summit in Pittsburgh (USA), some guys from Carnegie Mellon University thought it would be a nice chance to show the whole world how they are concerned about current geopolitical, environmental, and economical problems.

They are not only willing to ban mundane and grievous methods from the evolutionary computation outer space such as genetic algorithms, but they also want to support vector machines, free variables, and start a brigade against Bayesian discrimination! No duality gaps! No greedy search! No power laws! And all the hails for a world in which anyone will freely do a safer data minning!

Evolving Selling Rates Through Simulated Evolution

There's an interesting article in E-Commerce Times on the use of innovations to improve sales. The author states that artificial intelligence techniques -- such as genetic algorithms, expert systems, and fuzzy systems -- could help in such an endevour. An excerpt:

"Every customer touch point will capture more information that the system can then analyze to increase the chances of making decisions on how to sell more to customers. The software cloud will be using concepts like expert systems, genetic algorithms, and fuzzy logic to make recommendations on how the business will sell more, what to sell where, who to sell to -- and it will modify itself to be able to accomplish these optimizations."

I think big vendors, such as Amazon.com, are already using those AI stuffs.