[ Pobierz całość w formacie PDF ]
.At mutation rates of one per four generations, under selection for smallsizes, creatures will optimize to a genome size in the 22 to 30 instruction size range withinas little as 300 million instructions of elapsed time.Each of these runs will reach a localoptima which it evidently cannot escape from, although it may not be the global optima.4.2 INCREASING COMPLEXITYThe unrolled loop (section 3.1.1.8) is an example of the ability of evolution to produce anincrease in complexity, gradually over a long period of time.The interesting thing about theloop unrolling optimization technique is that it requires more complex code.The resultingcreature has a genome size of 36, compared to its ancestor of size 80, yet it has packed amuch more complex algorithm into less than half the space (Appendix E).This is a classic example of intricate design in evolution.One wonders how it could havearisen through random bit flips, as every component of the code must be in place in order forthe algorithm to function.Yet the code includes a classic mix of apparent intelligent design,and the chaotic hand of evolution.The optimization technique is a very clever one inventedby humans, yet it is implemented in a mixed up but functional style that no human woulduse (unless perhaps very intoxicated).4.3 EMERGENCEThe physical environment presented by the simulator is quite simple, consisting of theenergy resource (CPU time) doled out rather uniformly by the time slicer, and memoryspace which is completely uniform and always available.In light of the nature of the physicalenvironment, the implicit fitness function would presumably favor the evolution of creatureswhich are able to replicate with less CPU time, and this does in fact occur.However, much ofthe evolution in the system consists of the creatures discovering ways to exploit one-another.The creatures invent their own fitness functions through adaptation to their biotic ( living )environment.These ecological interactions are not programmed into the system, but emergespontaneously as the creatures discover each other and invent their own games.In the Tierran world, creatures which initially do not interact, discover means to exploitone another, and in response, means to avoid exploitation.The original fitness landscapeof the ancestor consists only of the efficiency parameters of the replication algorithm, in thecontext of the properties of the reaper and slicer queues.When by chance, genotypes appearthat exploit other creatures, selection acts to perfect the mechanisms of exploitation, andmechanisms of defense to that exploitation.The original fitness landscape was based only on20 adaptations of the organism to its physical environment.The new fitness landscape retainsthose features, but adds to it adaptations to the biotic environment, the other creatures.Because the fitness landscape includes an ever increasing realm of adaptations to other crea-tures which are themselves evolving, it can facilitate an auto-catalytic increase in complexityand diversity of organisms.Evolutionary theory suggests that adaptation to the biotic environment (other organisms)rather than to the physical environment is the primary force driving the auto-catalyticdiversification of organisms ([34]).It is encouraging to discover that the process has alreadybegun in the Tierran world.It is worth noting that the results presented here are basedon evolution of the first creature that I designed, written in the first instruction set that Idesigned.Comparison to the creatures that have evolved shows that the one I designed isnot a particularly clever one.Also, the instruction set that the creatures are based on iscertainly not very powerful (apart from those special features incorporated to enhance itsevolvability).It would appear then that it is rather easy to create life.Evidently, virtual lifeis out there, waiting for us to provide environments in which it may evolve.4.4 SYNTHETIC BIOLOGYOne of the most uncanny of evolutionary phenomena is the ecological convergence of biota liv-ing on different continents or in different epochs.When a lineage of organisms undergoes anadaptive radiation (diversification), it leads to an array of relatively stable ecological forms.The specific ecological forms are often recognizable from lineage to lineage.For exampleamong dinosaurs, the Pterosaur, Triceratops, Tyrannosaurus and Ichthyosaur are ecolog-ical parallels respectively, to the bat, rhinoceros, lion and porpoise of modern mammals.Similarly, among modern placental mammals, the gray wolf, flying squirrel, great anteaterand common mole are ecological parallels respectively, to the Tasmanian wolf, honey glider,banded anteater and marsupial mole of the marsupial mammals of Australia.Given these evidently powerful convergent forces, it should perhaps not be surprisingthat as adaptive radiations proceed among digital organisms, we encounter recognizableecological forms, in spite of the fundamentally distinct physics and chemistry on which theyare based.Ideally, comparisons should be made among organisms of comparable complexity.It may not be appropriate to compare viruses to mammals.Unfortunately, the organiccreatures most comparable to digital organisms, the RNA creatures, are no longer with us [ Pobierz całość w formacie PDF ]
zanotowane.pl doc.pisz.pl pdf.pisz.pl centka.pev.pl
.At mutation rates of one per four generations, under selection for smallsizes, creatures will optimize to a genome size in the 22 to 30 instruction size range withinas little as 300 million instructions of elapsed time.Each of these runs will reach a localoptima which it evidently cannot escape from, although it may not be the global optima.4.2 INCREASING COMPLEXITYThe unrolled loop (section 3.1.1.8) is an example of the ability of evolution to produce anincrease in complexity, gradually over a long period of time.The interesting thing about theloop unrolling optimization technique is that it requires more complex code.The resultingcreature has a genome size of 36, compared to its ancestor of size 80, yet it has packed amuch more complex algorithm into less than half the space (Appendix E).This is a classic example of intricate design in evolution.One wonders how it could havearisen through random bit flips, as every component of the code must be in place in order forthe algorithm to function.Yet the code includes a classic mix of apparent intelligent design,and the chaotic hand of evolution.The optimization technique is a very clever one inventedby humans, yet it is implemented in a mixed up but functional style that no human woulduse (unless perhaps very intoxicated).4.3 EMERGENCEThe physical environment presented by the simulator is quite simple, consisting of theenergy resource (CPU time) doled out rather uniformly by the time slicer, and memoryspace which is completely uniform and always available.In light of the nature of the physicalenvironment, the implicit fitness function would presumably favor the evolution of creatureswhich are able to replicate with less CPU time, and this does in fact occur.However, much ofthe evolution in the system consists of the creatures discovering ways to exploit one-another.The creatures invent their own fitness functions through adaptation to their biotic ( living )environment.These ecological interactions are not programmed into the system, but emergespontaneously as the creatures discover each other and invent their own games.In the Tierran world, creatures which initially do not interact, discover means to exploitone another, and in response, means to avoid exploitation.The original fitness landscapeof the ancestor consists only of the efficiency parameters of the replication algorithm, in thecontext of the properties of the reaper and slicer queues.When by chance, genotypes appearthat exploit other creatures, selection acts to perfect the mechanisms of exploitation, andmechanisms of defense to that exploitation.The original fitness landscape was based only on20 adaptations of the organism to its physical environment.The new fitness landscape retainsthose features, but adds to it adaptations to the biotic environment, the other creatures.Because the fitness landscape includes an ever increasing realm of adaptations to other crea-tures which are themselves evolving, it can facilitate an auto-catalytic increase in complexityand diversity of organisms.Evolutionary theory suggests that adaptation to the biotic environment (other organisms)rather than to the physical environment is the primary force driving the auto-catalyticdiversification of organisms ([34]).It is encouraging to discover that the process has alreadybegun in the Tierran world.It is worth noting that the results presented here are basedon evolution of the first creature that I designed, written in the first instruction set that Idesigned.Comparison to the creatures that have evolved shows that the one I designed isnot a particularly clever one.Also, the instruction set that the creatures are based on iscertainly not very powerful (apart from those special features incorporated to enhance itsevolvability).It would appear then that it is rather easy to create life.Evidently, virtual lifeis out there, waiting for us to provide environments in which it may evolve.4.4 SYNTHETIC BIOLOGYOne of the most uncanny of evolutionary phenomena is the ecological convergence of biota liv-ing on different continents or in different epochs.When a lineage of organisms undergoes anadaptive radiation (diversification), it leads to an array of relatively stable ecological forms.The specific ecological forms are often recognizable from lineage to lineage.For exampleamong dinosaurs, the Pterosaur, Triceratops, Tyrannosaurus and Ichthyosaur are ecolog-ical parallels respectively, to the bat, rhinoceros, lion and porpoise of modern mammals.Similarly, among modern placental mammals, the gray wolf, flying squirrel, great anteaterand common mole are ecological parallels respectively, to the Tasmanian wolf, honey glider,banded anteater and marsupial mole of the marsupial mammals of Australia.Given these evidently powerful convergent forces, it should perhaps not be surprisingthat as adaptive radiations proceed among digital organisms, we encounter recognizableecological forms, in spite of the fundamentally distinct physics and chemistry on which theyare based.Ideally, comparisons should be made among organisms of comparable complexity.It may not be appropriate to compare viruses to mammals.Unfortunately, the organiccreatures most comparable to digital organisms, the RNA creatures, are no longer with us [ Pobierz całość w formacie PDF ]