Gated Recurrent Unit (GRU)
The Gated Recurrent Unit (GRU) is a type of Recurrent Neural Network (RNN) that, in certain cases, has advantages over long short term memory (LSTM). GRU uses less memory and is faster than LSTM, however, LSTM is more accurate when using datasets with longer sequences.
Also, GRUs address the vanishing gradient problem (values used to update network weights) from which vanilla recurrent neural networks suffer. If the grading shrinks over time as it back propagates, it may become too small to affect learning, thus making the neural net untrainable.
If layer in a neural net can’t learn, RNN’s can essentially “forget” longer sequences.
GRUs solve this problem through the use of two gates, the update gate and reset gate. These gates decide what information is allowed through to the output and can be trained to retain information from farther back. This allows it to pass relevant information down a chain of events to make better predictions.