Uncommon Descent Serving The Intelligent Design Community

Putting Richard Dawkins’s evolution claims to a computer science test

Share
Facebook
Twitter
LinkedIn
Flipboard
Print
Email

Eric Holloway asks, “Do bugs really show no evidence of design? Computer engineers can test that…

Dawkins does not think that there really is any design in nature. His solution is Darwinian evolution, by which inventions emerge by one tiny incremental step at a time in response to natural selection (survival of the fittest).

Can we test that?

As it happens, we can simulate this process on a computer, using evolutionary algorithms. Do we get marvels of mechanical and material engineering, such as bioluminescence and waterproof soccer balls?

Engineers have tried Darwinian evolution for the past decade and a half at a competition called the Humies, in which new human inventions, produced by evolutionary algorithms, compete. However, if you scan the list of past winners, you’ll note a pattern. All the awards are for what is called “parameter tuning.” The actual hard creative work of coming up with the overall blueprint of the invention has already been done by a human engineer and encoded as an objective function for the evolutionary algorithm to optimize. The evolutionary algorithm has already been told where to go.

When we apply Dawkins’s “blind evolution” explanation of the abilities exhibited by insects to the real world engineering, all we get is fancy knob twiddlers.

Eric Holloway, “Technology? Hey, bugs invented it all before us!” at Mind Matters News

See also: Human, mouse, and fly brains all use the same basic mechanisms. The study of brains in recent decades has yielded a very different picture from the patterns we might have expected.

and

Why do many scientists see cells as intelligent? Bacteria appear to show intelligent behavior. But what about individual cells in our bodies?

Comments
Algorithms do not create their own goals. All goals are implied in the human-programmed rules.mike1962
October 14, 2020
October
10
Oct
14
14
2020
05:45 PM
5
05
45
PM
PDT
So called evolutionary algorithms are simply a computerised version of the trial and error solutions that engineers used on paper for centuries. The only real difference is that computers allow many more trials than before.aarceng
October 14, 2020
October
10
Oct
14
14
2020
01:11 PM
1
01
11
PM
PDT
this article mentions so called "evolutionary algorithms" Through the years, I debated lots of Darwinian clowns. Some of them, sometimes mentioned "evolutionary algorithms" So i had a look, what these Darwinian clowns talk about... I found out, that evolutionary algorithm is just a fancy word for optimization algorithm. This is nothing new. For some very limited applications, some optimization using such an algorithm can be useful... In fact, these algorithms are useless, no engineer ever use it when creating/developing something new... Here is a proof, a time ago I did my own survey, i have contacted top-notch engineering companies, their engineers, who deal with developing humanoid robots ... NONE of them ever used evolutionary/genetic algorithms, here is my communication: A company from Spain: "Our engineers do not use GA/EA in our company and due to that, I can not help you with much more. Here is another link where you can have some information. Wishing you all the best." A company from Italy: "Engineer 20+ years, No, we haven’t used genetic/evolutionary algorithms." (this guy was an engineer for 20+ years) Another company/university from Italy: ""Unfortunately I am an academic so I cannot give you an answer from the point of view of a person from industry but I will try to answer anyway. From the research perspective there is a considerable activity on GE and I see an increasing interest in EA (as a precursor of embodied/interactive learning). From my personal experience I agree with you in that I don’t see a lot of activity around this topics in the robotics industry." " A company from France: "No we haven’t used such technique in real cases. It seems a great tool allowing the exploration of different design/approach. Yet, I have only seen such uses in research applications atm." and so on ... in other words, no engineer ever used evolutionary algorithms IN REAL LIFE :)))))))) Darwinian clowns ... HELLO ????martin_r
October 14, 2020
October
10
Oct
14
14
2020
10:14 AM
10
10
14
AM
PDT
Yup. Parameter tuning isn't new. Mechanical devices have been doing it for 2000 years, since the invention of the centrifugal governor. When a program or machine invents negative feedback ON ITS OWN, I might be impressed. So far negative feedback is solely produced by life, or by devices intelligently made by humans.polistra
October 13, 2020
October
10
Oct
13
13
2020
07:15 PM
7
07
15
PM
PDT

Leave a Reply