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.

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.

Auf Wiedersehen!

Hi, Friends! :)

So, the Genetic Argonaut blog will soon enter some kind of "stagnation" (but not in a local optimum, since this blog has self-adaptation mechanisms!), because I am planning to begin my final project at university, something that has to do with neural networks. Then, the update of this blog will be sporadically, but not so "frequent" as it has been so far.

However, after some months I will be back again! Just wait!

Auf Wiedersehen!

Marcelo