4 ms·

Does information theory support design in nature?

2 points by yters 8y ago

bediger4000 8y ago
The title of the article seems... disingenuous. First, the article doesn't answer that question. Second, the title and the article assume that "design" appears in nature. The article doesn't support this claim.
- yters 8y ago
  The referenced article has more details. Also the comments have more details. https://mindmatters.ai/2018/10/does-information-theory-support-design-in-nature/ https://mindmatters.ai/2018/10/does-information-theory-suppo... I'm happy to answer any questions you have.
  - bediger4000 8y ago
    Sure. Why is the title of the article so disingenuous? What design do people think they see in nature? By what method can I, without referring to someone else's authority, decide that a particular part of nature is "designed"? Without an explicit criteria for "design", the whole thing is just arguing over opinions about what is designed and what isn't. That is, it would become just like Behe's "irreducible complexity". No definite criteria for "irreducible" and no measurement for "complexity". While you're at it, point me to some PDFs of refereed papers written by Dembski. I'm a smart guy, I can read primary sources. The article doesn't point to any, and a quick duckduckgo for "dembski no free lunch" gets me web pages that provide a critique of Dembski's book (and not on information theoretic grounds, either), and his book on Amazon, but nothing I can print and read and digest. No PDFs. How does Dembski wave off actual examples of artificial evolution? Tierra (http://life.ou.edu/tierra/ http://life.ou.edu/tierra/) seems to evolve little digital pseudo-life that does a specific task. Here's another example: http://www.genetic-programming.com/jkpdf/alife1990.pdf http://www.genetic-programming.com/jkpdf/alife1990.pdf There are more "folksy" reports, too: https://www.allegro.cc/forums/thread/293256 https://www.allegro.cc/forums/thread/293256 . There's an entire field, genetic algorithms/genetic programming, that actually ends up at least some of the time, evolving a solution to the problem. Adrian Thompson's FPGA experiment (https://www.researchgate.net/profile/Adrian_Thompson5/publication/285668389_Exploring_beyond_the_scope_of_human_design_Automatic_generation_of_FPGA_configurations_through_artificial_evolution/links/56f0a8bd08ae584badc9389e/Exploring-beyond-the-scope-of-human-design-Automatic-generation-of-FPGA-configurations-through-artificial-evolution.pdf https://www.researchgate.net/profile/Adrian_Thompson5/public...) is pretty famous. Here's a distillation of the above paper with a few interesting editorial comments: https://static.aminer.org/pdf/PDF/000/308/779/an_evolved_circuit_intrinsic_in_silicon_entwined_with_physics.pdf https://static.aminer.org/pdf/PDF/000/308/779/an_evolved_cir... How does Dembski explain how every system that attempts to "evolve" something by successive generations that inherit and have minor changes (https://arxiv.org/abs/1803.03453 https://arxiv.org/abs/1803.03453, which is a summary of many experiments) end up doing something that's undeniably evolution?
    - yters 8y ago
      Sorry about the title, I should have linked to the primary article instead of the one with the discussion. I submitted again with a direct link this time. https://news.ycombinator.com/item?id=18449438 https://news.ycombinator.com/item?id=18449438 The design detection is essentially a hypothesis test, and doesn't depend on anyone's authority. I'm happy to answer questions on how this works. The basic idea is Shannon information says all bitstrings of length N are equally likely, but Kolmogorov complexity says the compressible bitstrings are less likely than the incompressible bitstrings. Therefore, when we see a highly compressible bitstring we infer some cause other than chance. E.g. if someone flips 100 heads in a row, you would be justified in thinking the coin is not fair. Dembski's work is a generalization of this observation. You can find a list of Dembski's refereed publications with google scholar. Here is a list of some that I found. A couple address your question regarding evolutionary algorithms. - https://ieeexplore.ieee.org/abstract/document/5204206 https://ieeexplore.ieee.org/abstract/document/5204206 - https://ieeexplore.ieee.org/abstract/document/5345941 https://ieeexplore.ieee.org/abstract/document/5345941 - https://link.springer.com/article/10.1007%2FBF01046100?LI=true https://link.springer.com/article/10.1007%2FBF01046100?LI=tr... - https://ieeexplore.ieee.org/abstract/document/5442816 https://ieeexplore.ieee.org/abstract/document/5442816
      - bediger4000 8y ago
        How do we go from some life thing, designed or not, to a bitstring? Don't say use DNA: representations of it are notoriously compressible, starting with the kind-of-error-correcting encoding (https://en.wikipedia.org/wiki/DNA_codon_table https://en.wikipedia.org/wiki/DNA_codon_table) of amino acid production. And thank you for the Avida ref. I could not remember its name, nor could I find it in google. Based on the abstracts of the papers, I can safely say that Dembski is mistaking "search with a goal" with evolution, which seems pretty un-directed almost by definition. His papers are probably true, and may be worth reading, but they're pretty obviously circular if you try to connect them to "design": "Computer search often uses an oracle to determine the value of a proposed problem solution. Information is extracted from the oracle using repeated queries." That's true: evolutionary computing has a human-decided goal, and therefore each generation is in essence a query of an oracle. Where's the oracle in terrestrial biology? That "oracle" is the "designer", but if the "oracle" is just whatever stays alive long enough to reproduce, his basic assumption of search isn't true: there's no goal.

2 points by yters 8y ago

The title of the article seems... disingenuous. First, the article doesn't answer that question. Second, the title and the article assume that "design" appears in nature. The article doesn't support this claim.

yters 8y ago
The referenced article has more details. Also the comments have more details. https://mindmatters.ai/2018/10/does-information-theory-support-design-in-nature/ https://mindmatters.ai/2018/10/does-information-theory-suppo... I'm happy to answer any questions you have.
- bediger4000 8y ago
  Sure. Why is the title of the article so disingenuous? What design do people think they see in nature? By what method can I, without referring to someone else's authority, decide that a particular part of nature is "designed"? Without an explicit criteria for "design", the whole thing is just arguing over opinions about what is designed and what isn't. That is, it would become just like Behe's "irreducible complexity". No definite criteria for "irreducible" and no measurement for "complexity". While you're at it, point me to some PDFs of refereed papers written by Dembski. I'm a smart guy, I can read primary sources. The article doesn't point to any, and a quick duckduckgo for "dembski no free lunch" gets me web pages that provide a critique of Dembski's book (and not on information theoretic grounds, either), and his book on Amazon, but nothing I can print and read and digest. No PDFs. How does Dembski wave off actual examples of artificial evolution? Tierra (http://life.ou.edu/tierra/ http://life.ou.edu/tierra/) seems to evolve little digital pseudo-life that does a specific task. Here's another example: http://www.genetic-programming.com/jkpdf/alife1990.pdf http://www.genetic-programming.com/jkpdf/alife1990.pdf There are more "folksy" reports, too: https://www.allegro.cc/forums/thread/293256 https://www.allegro.cc/forums/thread/293256 . There's an entire field, genetic algorithms/genetic programming, that actually ends up at least some of the time, evolving a solution to the problem. Adrian Thompson's FPGA experiment (https://www.researchgate.net/profile/Adrian_Thompson5/publication/285668389_Exploring_beyond_the_scope_of_human_design_Automatic_generation_of_FPGA_configurations_through_artificial_evolution/links/56f0a8bd08ae584badc9389e/Exploring-beyond-the-scope-of-human-design-Automatic-generation-of-FPGA-configurations-through-artificial-evolution.pdf https://www.researchgate.net/profile/Adrian_Thompson5/public...) is pretty famous. Here's a distillation of the above paper with a few interesting editorial comments: https://static.aminer.org/pdf/PDF/000/308/779/an_evolved_circuit_intrinsic_in_silicon_entwined_with_physics.pdf https://static.aminer.org/pdf/PDF/000/308/779/an_evolved_cir... How does Dembski explain how every system that attempts to "evolve" something by successive generations that inherit and have minor changes (https://arxiv.org/abs/1803.03453 https://arxiv.org/abs/1803.03453, which is a summary of many experiments) end up doing something that's undeniably evolution?
  - yters 8y ago
    Sorry about the title, I should have linked to the primary article instead of the one with the discussion. I submitted again with a direct link this time. https://news.ycombinator.com/item?id=18449438 https://news.ycombinator.com/item?id=18449438 The design detection is essentially a hypothesis test, and doesn't depend on anyone's authority. I'm happy to answer questions on how this works. The basic idea is Shannon information says all bitstrings of length N are equally likely, but Kolmogorov complexity says the compressible bitstrings are less likely than the incompressible bitstrings. Therefore, when we see a highly compressible bitstring we infer some cause other than chance. E.g. if someone flips 100 heads in a row, you would be justified in thinking the coin is not fair. Dembski's work is a generalization of this observation. You can find a list of Dembski's refereed publications with google scholar. Here is a list of some that I found. A couple address your question regarding evolutionary algorithms. - https://ieeexplore.ieee.org/abstract/document/5204206 https://ieeexplore.ieee.org/abstract/document/5204206 - https://ieeexplore.ieee.org/abstract/document/5345941 https://ieeexplore.ieee.org/abstract/document/5345941 - https://link.springer.com/article/10.1007%2FBF01046100?LI=true https://link.springer.com/article/10.1007%2FBF01046100?LI=tr... - https://ieeexplore.ieee.org/abstract/document/5442816 https://ieeexplore.ieee.org/abstract/document/5442816
    - bediger4000 8y ago
      How do we go from some life thing, designed or not, to a bitstring? Don't say use DNA: representations of it are notoriously compressible, starting with the kind-of-error-correcting encoding (https://en.wikipedia.org/wiki/DNA_codon_table https://en.wikipedia.org/wiki/DNA_codon_table) of amino acid production. And thank you for the Avida ref. I could not remember its name, nor could I find it in google. Based on the abstracts of the papers, I can safely say that Dembski is mistaking "search with a goal" with evolution, which seems pretty un-directed almost by definition. His papers are probably true, and may be worth reading, but they're pretty obviously circular if you try to connect them to "design": "Computer search often uses an oracle to determine the value of a proposed problem solution. Information is extracted from the oracle using repeated queries." That's true: evolutionary computing has a human-decided goal, and therefore each generation is in essence a query of an oracle. Where's the oracle in terrestrial biology? That "oracle" is the "designer", but if the "oracle" is just whatever stays alive long enough to reproduce, his basic assumption of search isn't true: there's no goal.