While talking about various and sundry methods of simulating natural processes like evolution in order to solve problems, the concept of a local minimum has cropped up from time to time. Today, we take a closer look at it, and some of the ways to escape.
In Darwin’s original theory, natural selection was considered to be driven by competition. This presents a problem. If it’s all about survival of the fittest, and out competing all others, why is there so much cooperation in nature? In the 1980/s, a tournament was held in which competitors sent in computer programs that would play something called “The Iterated Prisoners’ Dilemma.” The program that did the best most consistently, was one of the simplest: “Tit for Tat.” It was, and still is, an example of how things like symbiosis and cooperation could evolve.
Here are a couple of articles on “Tit for Tat,” and “The Prisoners’ Dilemma.”
Today we take a look at my favorite evolutionary approach to making computers solve problems. This one has, now and again, produced results that are competitive with what humans can do.
Here’s a link to a page about Genetic Programming, including some of the human competitive results.
Your DNA has most of the information needed to make you you. The information is encoded through four different chemicals that act as a sort of alphabet, spelling out the language of life. In the 1960/s a similar approach was adopted to solve complex problems.
Here’s a tutorial on genetic algorithms and how to use them.
In 1963, a couple of students were working in a wind-tunnel, attempting to find a way to figure out how to design surfaces to produce a given airflow. That was, and still is, a difficult problem. There are no simple formula. They decided to see if they could use the power of evolution to solve the problem, and it worked.
Here’s a link to the relevant section of “A Hitchhiker’s guide to Evolutionary Computation.”
I mentioned evolutionary programming in this episode, but don’t provide much detail. For those who are interested, here’s a link that provides an introduction to the topic.
“
To suppose that the eye, with all its inimitable contrivances… could have been formed by natural selection, seems, I freely confess, absurd in the highest possible degree… Yet reason tells me, that if numerous gradations from a perfect and complex eye to one very imperfect and simple, each grade being useful to its possessor, can be shown to exist… and if any variation or modification in the organ be ever useful to an animal under changing conditions of life, then the difficulty of believing that a perfect and complex eye could be formed by natural selection, though insuperable by our imagination, can hardly be considered a real1.
“
Charles Darwin (1809–1882)
Darwin included a section in his book, “On the Origin of Species,” called, “Problems With the Theory.” It included the Cambrian explosion, covered in the previous episode, and the development of complex organs like the eye. Today, using things we’ve learned since Darwin’s time, we look at how eyes evolved.
Here are a number of articles with further information.
Between 520 and 550 million years ago, a sudden explosion of animal types appear in the fossil record. This example of rapid evolution is known as the Cambrian explosion. Theories of how and why it occurred range from the notion that it didn’t happen at all, to a spike in oxygen levels, to the advent of the sense of vision.
Here’s an article that includes an animation of some of the oddball animals that appeared and disappeared during the Cambrian period.
In today’s rather short episode, we talk about the first creatures to have developed a centralized nervous system, though not a central nervous system as of yet. It was a simple worm like creature, with a nerve cord running along the length of its body, and an extra-large bundle of nerves toward its mouth. For worms and other invertebrate animals, like crabs, lobsters, octopuses, squid, slugs and snails, the nerve cord runs along the belly of the creature. For what would become vertebrates, including us, the main nerve cord runs along the back. Apparently, for reasons unknown, some of the worms flipped over, and decided to live their lives upside-down.
This month has included some unusually long episodes. To leave room for the upcoming Halloween special, this and the next couple of episodes are unusually short. Check back on the 31st for, “Be Afraid.: when reason can get you killed.”
b
Darwin originally published “On the Origins of Species” in 1859. At the time, the mechanism of inheritance wasn’t well understood. Inheritance and the implication that lifeforms could change over time stood in contrast to the popular notion that God had created all things in perfect and unchanging forms. His theory only considered traits that come from parents, in the case of sexual reproduction, and mutation, in asexual reproduction. It turns out that organisms, especially single celled forms, can and do snag genetic information from other organisms that they aren’t related to in the least. Today we look at some of the mechanisms of horizontal gene transfer, and some examples.
Bacteria are very successful. They’ve been around for billions of years, as compared to hundreds of millions for multicellular creatures. They have survived mass extinctions that wiped out things like the dinosaurs and others. Today, we look at salpingoeca rosetta, which can live as either a single celled creature, or in a multicellular colonial form. We compare that to the experimentally evolved multicellular colonies of yeast we introduced in the previous episode, episode 72; and examine how the multicellular creatures came about, and how and why they are no longer able to return to a single celled lifestyle.
Here’s an article about work by Nicole King, that introduced me to salpingoeca rosetta.