RNN stands for recurrent neural network and is a type of artificial neural network (ANN) which is a non feed-forward neural network (FNN). This means that an RNN accommodates for information flow from left to right but also from right to left. This is because an RNN typically features one or more memory cells inside it. A memory cell is an algorithm which stores data values of the current and the previous point in time. There can be simple memory cells as well as long short-term memory cells (LSTM) and Gated Recurrent Unit (GRU) cells. GRU cells are a simplified version of LSTM cells. RNN are most commonly used in Natural Language Processing (NLP) tasks.
