Robots have shown altruism in the process of evolution
Evolutionary biologists from the University of Lausanne (Switzerland) were able to prove that altruism in living beings arose as a result of natural selection, that is, it is an innate quality necessary for the survival of relatives. They showed it by the example of robots, whose software was modified using evolutionary algorithms.
As Hamilton ’s rule says, it can be beneficial for egoistic individuals to sacrifice their time and resources for the sake of their relatives, because it makes sense in terms of the preservation of genes. Hamilton for the first time gave altruism a rational explanation from the point of view of evolution and derived a formula according to which altruism should manifest itself in kin-selection .
The fundamental principle of kin-selection also extends to artificial life forms, that is, robots. Keller's group from the University of Lausanne was able to prove not only the presence of altruism among robots, but precisely the emergence of it as a result of natural selection, when individuals with more common genes showed greater altruism.
')
In this experiment, robots with infrared sensors were programmed to search for disks (“food”), which had to be pushed to a certain place. At the end of the life of each generation, the computer “genes” of successful robots intermingled and gave rise to a new generation of robots, and the genomes of the unlucky individuals were destroyed. Experience has shown that the degree of altruism in robots fully coincides with Hamilton’s calculations.
In groups of unrelated robots (with a small number of common genes), individuals tried to seek and push small disks, which is an ideal strategy in terms of individual survival. In groups of robots with a lot of common code, robots more often helped others push large disks.
As Hamilton ’s rule says, it can be beneficial for egoistic individuals to sacrifice their time and resources for the sake of their relatives, because it makes sense in terms of the preservation of genes. Hamilton for the first time gave altruism a rational explanation from the point of view of evolution and derived a formula according to which altruism should manifest itself in kin-selection .
The fundamental principle of kin-selection also extends to artificial life forms, that is, robots. Keller's group from the University of Lausanne was able to prove not only the presence of altruism among robots, but precisely the emergence of it as a result of natural selection, when individuals with more common genes showed greater altruism.
')
In this experiment, robots with infrared sensors were programmed to search for disks (“food”), which had to be pushed to a certain place. At the end of the life of each generation, the computer “genes” of successful robots intermingled and gave rise to a new generation of robots, and the genomes of the unlucky individuals were destroyed. Experience has shown that the degree of altruism in robots fully coincides with Hamilton’s calculations.
In groups of unrelated robots (with a small number of common genes), individuals tried to seek and push small disks, which is an ideal strategy in terms of individual survival. In groups of robots with a lot of common code, robots more often helped others push large disks.
Source: https://habr.com/ru/post/118764/
All Articles