Content deleted Content added
m Modified GA to EA. GAs are just one specific type of EA, but all EAs are sensibly part of SoftComp (see https://www.journals.elsevier.com/applied-soft-computing) |
No edit summary Tag: Reverted |
||
Line 8:
including [[neural network]]s, [[fuzzy logic]], and [[evolutionary algorithm]]s.<ref>
{{Citation|last=Shukla|first=K. K.|title=CHAPTER 17 - Soft Computing Paradigms for Artificial Vision|date=2000-01-01|url=https://www.sciencedirect.com/science/article/pii/B9780126464900500202|work=Soft Computing and Intelligent Systems|pages=405–417|editor-last=Sinha|editor-first=NARESH K.|series=Academic Press Series in Engineering|place=San Diego|publisher=Academic Press|language=en|isbn=978-0-12-646490-0|access-date=2021-02-24|editor2-last=Gupta|editor2-first=MADAN M.}}
</ref>Among these, neural networks are usually used for prediction. The basic structure of multilayer perception neural network consists of the input layer, middle hidden layer and the output layer, with the product of input factors (ai) and weights (wij) fed to summing junctions with neurons bias (bj).<ref>Sustainable Construction Safety Knowledge Sharing: A Partial Least Square-Structural Equation Modeling and A Feedforward Neural Network Approach. Sustainability 2019, 11, 5831. https://doi.org/10.3390/su11205831</ref>
These algorithms are tolerant of imprecision, uncertainty, partial truth and approximation.
It is contrasted with '''hard computing''': algorithms which find provably correct and [[optimal]] solutions to problems.
| |||