Chapter 121 19.10 Mathematical principles of natural selection

For artificial evolution to be as creative as natural evolution, we must either provide it with infinite time that we cannot, or enhance it by borrowing from natural evolution's more creative elements, if any.But at least artificial evolution can help us elucidate the true character of the evolution of life on Earth; something neither existing observations nor historical fossils can do. I'm not at all concerned that evolutionary theory might be taken over by post-Darwinists who don't have degrees in biology.Artificial evolution has long taught us an important lesson—evolution is not a biological process.It integrates technical, mathematical, informational and biological processes.It could almost be said that evolution is a law of physics that applies to all populations, whether they have genes or not.

The most unacceptable part of Darwin's theory of natural selection is its inevitability.The conditions of natural selection are very special, but once these conditions are met, natural selection will inevitably occur! Natural selection can only occur among populations or clusters of things.This is actually a disorder phenomenon that occurs in space and time.The population involved in this process must have the following characteristics: (1) there are some changes in characteristics among individuals; (2) these characteristics bring about some differences in the fertility, fecundity or survival ability of individuals; (3) These traits can be passed from parent to offspring in some way.If these conditions are met, natural selection will inevitably occur, just like 6 must be followed by 7, or a coin must have two sides.As evolutionary theorist John Endel put it, "Natural selection probably shouldn't be called a law of biology. It happens not because of biology, but because of probability theory."

But natural selection is not evolution, nor is evolution the same as natural selection.Just as arithmetic is not mathematics, mathematics is not arithmetic.Of course, you can claim that mathematics is really just a combination of addition.Subtraction is the inverse of addition, multiplication is consecutive addition, and all complex functions based on these operations are just extensions of addition.This is somewhat similar to the logic of the neo-Darwinists: all evolution is an extension of the combination of natural selection.While there is a grain of truth to this, it blocks our understanding and acceptance of more complex things.Multiplication is indeed a kind of addition, but out of this shortcut emerges a whole new power that we would never have mastered if we thought of multiplication only as a repetition of addition.Be content with addition only, and you will never get E=mc2.

I believe there is a mathematics of life.Natural selection is perhaps this mathematical addition.To adequately explain the origin of life, trends in complexity, and the emergence of intelligence requires not just addition but a rich mathematics of complex functions that build upon each other.It requires a deeper evolution.Natural selection alone is not enough.To make a difference, more creative and productive processes must be incorporated.It must have more means than natural selection. Post-Darwinists have demonstrated that there is no single evolution driven by natural selection in one dimension.Evolution should have both breadth and depth.Deep evolution is the aggregation of multiple evolutions. It is a multi-faceted god and a thousand-armed creator. His methods of creation are various, and natural selection may be the most universal method among them.Deep evolution is made up of these many, many unrecognized evolutions, as if our minds were an inclusive society.Different evolutions operate on different scales, at different rhythms, and in different styles.Furthermore, the evolution of this mix has changed over time.Certain types of evolution were important to early proto-life, others are more responsible today, four billion years later.Some kind of evolution (natural selection) will appear everywhere, other evolution may just happen occasionally and play a specific role.This deep evolution of diversity, like intelligence, emerges from a certain dynamic community.

This heterogeneous nature of evolution also needs to be taken into account when we build artificial evolution to breed machines or software.I look forward to seeing the following characteristics in artificial evolution with openness and sustainable creativity (I believe these characteristics are also present in biological evolution, but artificial evolution will express these characteristics more significantly): Symbiosis - a convenient exchange of information to allow different evolutionary paths to come together Directed mutation - non-random mutation and direct communication and swapping mechanisms with the environment

Hopping – clustering of functions, hierarchies of control, modularization of components, and adaptation processes that change many properties simultaneously Self-organization - the developmental process of favoring a particular shape (such as four wheels) and making it a universal standard Artificial evolution cannot create everything.Although we can imagine many things exhaustively—and judging by the laws of physics and logic, they must also work—we cannot actually realize them due to the constraints of synthetic evolution itself. The post-Darwinists with their computers all day are subconsciously asking: Where is the limit of evolution?What can't evolution do?The limit of organism evolution may not be broken through, but its inclinations and limitations may contain the answers prepared for the geniuses who devote themselves to the study of evolution.In the wilderness of possible creatures, where are there unoccupied black holes?For this I can only quote that eccentric Abbotcher who said: "I am more concerned with the blank spaces, with the shapes that can be imagined but cannot be realized." In Lewontin's words: "Evolution cannot produce all things, but it can explain something."