China 
Africa 
Explore chapters and articles related to this topic
The Joy of Counterintuitivity
Published in Elizabeth Mossop, Sustainable Coastal Design and Planning, 2018
Fitness landscapes are complex adaptive systems that self-organize and adapt, in order to remain within their current state. The system only shifts to other attractors (alternative states) after a shock that drives the system out of its current state (e.g., due to significant change). Major adjustments are needed and, after the initial shock, the system will self-organize to achieve those adjustments. The process this system goes through can be represented in the form of a fitness landscape (Langton et al., 1992; Mitchell Waldrop, 1992; Cohen and Stewart, 1994). This fitness landscape includes positions considered favorable (the peaks) and less favorable (the valleys). A complex system tends to move across less favorable valleys to the highest possible position in the landscape, the attractor. At the peak, the adaptive capacity is highest, which allows the system to adapt more easily to changes in its environment. The design of a landscape should freely allow the dynamic play of elements that will enable that landscape's potential to re-shape and re-organize itself and seek a new stable situation.
Evolutionary Learning
Published in Stephen Marsland, Machine Learning, 2014
The fitness function can be seen as an oracle that takes a string as an argument and returns a value for that string. Together with the string encoding the fitness function forms the problem-specific part of the GA. It is worth thinking about what we want from our fitness function. Clearly, the best string should have the highest fitness, and the fitness should decrease as the strings do less well on the problem. In real evolution, the fitness landscape is not static: there is competition between different species, such as predators and prey, or medical cures for certain diseases, and so the measure of fitness changes over time. We’ll ignore that in the genetic algorithm.
Introduction — A Tutorial of Sorts
Published in Brian E. White, Toward Solving Complex Human Problems, 2020
Work in biology on “fitness landscapes” is an interesting illustration of competitive co-evolution [68]. A fitness landscape is based on the idea that the fitness of an organism is not dependent only on its intrinsic characteristics but also on its interaction with its environment. The term “landscape” comes from visualizing a geographical landscape of fitness “peaks,” where each peak represents an adaptive solution to a problem of optimising certain kinds of benefits to the species. The fitness landscape is most appropriately used where there is a clear single measure of the fitness of an entity, so may not always be useful in social sciences [50, pp. 53–54].
Technology fitness landscape for design innovation: a deep neural embedding approach based on patent data
Published in Journal of Engineering Design, 2022
The creation of the technology fitness landscape was inspired by Kauffman’s NK genome fitness landscape for assessing the evolution of genotypes. The genome fitness landscape comprises genotypes similar or dissimilar to each other to different degrees, and the height of an area corresponds to the fitness (or replication rate) of a particular genotype. In the genome fitness landscape, each genotype is composed of several nucleotides (DNA sequences, which can be viewed as 4-d vectors), and each position in a DNA sequence can be occupied by four alternative bases: adenine (A), thymine (T), guanine (G), and cytosine (C). By analogy, the technology domains in the total technology fitness landscape are similar to the genotypes. The domain features can be viewed as the nucleotides in the DNA sequence of the technological domain and are now characterized by our 32-dimensional vectors trained on multimodal data.