Editing Bio-inspired computing (section)

===Population Based Bio-Inspired Algorithms===
Bio-inspired computing, which work on a population of possible solutions in the context of [[evolutionary algorithm]]s or in the context of [[swarm intelligence]] algorithms, are subdivided into '''Population Based Bio-Inspired Algorithms''' (PBBIA).<ref>{{cite journal |last1=Farinati |first1=Davide |last2=Vanneschi |first2=Leonardo |title=A survey on dynamic populations in bio-inspired algorithms |journal=Genetic Programming and Evolvable Machines |date=December 2024 |volume=25 |issue=2 |doi=10.1007/s10710-024-09492-4|hdl=10362/170138 |hdl-access=free }}</ref> They include [[Evolutionary Algorithm]]s, [[Particle Swarm Optimization]], [[Ant colony optimization algorithms]] and [[Artificial bee colony algorithm]]s.

==== Virtual Insect Example ====
Bio-inspired computing can be used to train a virtual insect. The insect is trained to navigate in an unknown terrain for finding food equipped with six simple rules:

* turn right for target-and-obstacle left; 
* turn left for target-and-obstacle right; 
* turn left for target-left-obstacle-right; 
* turn right for target-right-obstacle-left;
* turn left for target-left without obstacle;
* turn right for target-right without obstacle.
The virtual insect controlled by the trained [[spiking neural network]] can find food after training in any unknown terrain.<ref name="Silvia_2013">{{cite book | author = Xu Z |author2=Ziye X |author3=Craig H |author4=Silvia F |title=52nd IEEE Conference on Decision and Control |chapter=Spike-based indirect training of a spiking neural network-controlled virtual insect | journal = IEEE Decision and Control | pages = 6798–6805 |date=Dec 2013 | doi = 10.1109/CDC.2013.6760966 | isbn = 978-1-4673-5717-3 |citeseerx=10.1.1.671.6351 |s2cid=13992150 }}</ref> After several generations of rule application it is usually the case that some forms of complex behaviour [[Emergence|emerge]]. Complexity gets built upon complexity until the result is something markedly complex, and quite often completely counterintuitive from what the original rules would be expected to produce (see [[complex system]]s). For this reason, when modeling the [[neural network (machine learning)|neural network]], it is necessary to accurately model an ''in vivo'' network, by live collection of "noise" coefficients that can be used to refine statistical inference and extrapolation as system complexity increases.<ref>{{cite web|url=http://www.duke.edu/~jme17/Joshua_E._Mendoza-Elias/Research_Interests.html#Neuroscience_-_Neural_Plasticity_in|title="Smart Vaccines" – The Shape of Things to Come|author=Joshua E. Mendoza|work=Research Interests|archiveurl=https://web.archive.org/web/20121114233853/http://people.duke.edu/~jme17/Joshua_E._Mendoza-Elias/Research_Interests.html|archivedate=November 14, 2012}}</ref>

Natural evolution is a good analogy to this method–the rules of evolution ([[Selection (biology)|selection]], [[Genetic recombination|recombination]]/reproduction, [[mutation]] and more recently [[transposition (genetics)|transposition]]) are in principle simple rules, yet over millions of years have produced remarkably complex organisms. A similar technique is used in [[genetic algorithm]]s.