Artificial Life and the likes

An old draft of my thoughts on artificial life and evolution, most of this is in my thoughts on philosophy, and the book reviews of Rachels and John Mainard Smith.

Artificial Life and the likes

Artificial life, artificial intelligence and evolution are all things that I find very interesting. I find the information given by the TV, SF films, and the like often badly misleading or plainly wrong (often because it either devilish, or imbibed of a romantic view of intelligence or man).

Evolution

In my view Evolution is a cornerstone of the scientific knowledge. I find it very depressing that it is still called "theory" of evolution and that in many places (as for example in the USA) its teaching is still contested. Obviously I think that everyone has the right to have his opinion and disagree (and in fact, science can never claim absolute truths), but in my view, this "theory" has more data that confirms it, that many scientific "facts".

If one has some "individuals" that are capable of some sort of "copying", where the copies have some traits of the originals, but are not identical to them, and the changes can influence the probability of creating further copies, then one has evolution. Now it is clear that for sexual reproduction this is the case.

All this thinking is independent of the details of inheritance, but to have a better theory, one can look at the molecular biology (DNA, genes,...), at experiments and at fossil traces.

Surely one of the most vocal defenders of the evolution theory is Dawkins, and even if I don't accept all his assertions his books are interesting and worthwhile to read (and contain many examples that validate the evolution theory). Very interesting to better understand the evolution is the excellent book Major transitions in Evolution by John Mainard Smith. In this book he looks at big transitions in the evolution (the origin of life, first cell, multicellular organisms, ...) and looks at how they could have happened.

Often the opposition to the evolution theory is not based on objective data, but stems from religious or ethical considerations. Unfortunately, many twisted ethical interpretations of the evolution had success just before the second world war. In my opinion, this stemmed mainly from the ignorance of the following (counterintuitive) facts about the evolution:

  1. The evolution is not about the survival of the strongest (if you don't define the strongest as the one that survives, making the definition circular and meaningless), in some situation the fittest can be the one that helps the others.
  2. The evolution does not necessarily strive against a "better" individual, and can be stuck in a local optimum or even go away from the optimum once reached (in some cases the advantage is so small that it get lost against the importance of maintaining the diversity in a changing environment.
Then there is the Hume guillotine: you can't derive what it should be from how a thing is (i.e. ethics from a fact).

The evolution theory explains how it is possible to have a complex structure/behavior/... that could not be happened by chance from nothing, to gradually evolve accumulating one small improvement over the other. In this case there is no designer, no goal, but there is a reason for the structure to be as it is, a reason that depends on the history there was to come to it.

I think that the evolution theory has ethical consequences, and Rachels in his book Nati dagli animali looks at them.

Artificial Intelligence

What is intelligence, if artificial intelligence is possible and how it would look like is an interesting topic. I have thought many times about it, but for now I will just point to Gödel, Escher, Bach, an eternal golden braid by R. Hofsteter; a book that looks at this problem from the perspective of mathematical logic.

Artificial Life

Tightly coupled to artificial intelligence is the topic of artificial life.

Ecas, was one of my ideas immediately after high school, that using something like Swarm to see emergent behaviors, my thinking was that gradients and locality are very important to be able to build emergent system features.

I never did pursue it seriously though.

Leave a comment

Comments are moderated. Your email address is neither published nor stored, only an md5 hash of it. Required fields are marked with *

Loading...