Editing Learning curve (section)

==In machine learning==
{{further|Learning curve (machine learning)}}
Plots relating performance to experience are widely used in [[machine learning]]. Performance is the error rate or accuracy of the [[learning]] system, while experience may be the number of training examples used for learning or the number of iterations used in [[optimizing]] the system model parameters.<ref>{{cite book|last1=Sammut|first1=Claude|title=Encyclopedia of Machine Learning|publisher=Springer|isbn=978-0-387-30768-8|page=578|url=https://books.google.com/books?id=i8hQhp1a62UC&q=neural+network+learning+curve&pg=PT604|edition=1st|editor=Webb, Geoffrey I. |date=2011}}</ref> The machine learning curve is useful for many purposes including comparing different algorithms,<ref>{{cite web|last=Madhavan|first=P.G.|title=A New Recurrent Neural Network Learning Algorithm for Time Series Prediction|url=http://www.jininnovation.com/RecurrentNN_JIntlSys_PG.pdf|work=Journal of Intelligent Systems|page=113, Fig. 3|volume=7|issue=1–2|date=1997}}</ref>  choosing model parameters during design,<ref>{{cite web|last=Singh|first=Anmol|date=2021|title=Machine learning for astronomy with scikit learning|url=https://sites.google.com/view/machinelearningforastronomy/home|publisher=Learning Curve My Personal Tutor}}</ref> adjusting optimization to improve convergence, and determining the amount of data used for training.<ref>{{cite journal|last=Meek|first=Christopher|author2=Thiesson, Bo |author3=Heckerman, David |title=The Learning-Curve Sampling Method Applied to Model-Based Clustering|journal=Journal of Machine Learning Research|date=Summer 2002|volume=2|issue=3|page=397|url=https://jmlr.csail.mit.edu/papers/volume2/meek02a/meek02a.pdf}}</ref>