olemus
ISO/IEC 22989:
neurovõrk, milles hetkekihti sisestatakse eelmise kihi ja eelmise töötlussammu väljastised
= neural network in which outputs from both the previous layer and the previous processing step are fed into the current layer

ISO/IEC 24029:
neurovõrk, mis säilitab mingi siseoleku, milles kodeeritakse see, mida neurovõrk on õppinud pärast sisendandmete mingi alamjärjendi töötlust
= neural network maintaining an internal state which encodes what the neural network has learned after processing a subsequence of the input data

ülevaateid
https://www.jeremyjordan.me/introduction-to-recurrent-neural-networks/

https://en.wikipedia.org/wiki/Recurrent_neural_network

http://imatge-upc.github.io/telecombcn-2016-dlcv/slides/D2L6-recurrent.pdf

https://web.eecs.umich.edu/~justincj/slides/eecs498/498_FA2019_lecture12.pdf

http://www.wildml.com/2015/09/recurrent-neural-networks-tutorial-part-1-introduction-to-rnns/

https://www.freecodecamp.org/news/the-ultimate-guide-to-recurrent-neural-networks-in-python/

https://analyticsindiamag.com/10-best-free-resources-to-learn-recurrent-neural-networks-rnns/