Long short-term memory. LSTM. Gated recurrent unit. GRU.

LSTM

Source:

https://en.m.wikipedia.org/wiki/Long_short-term_memory

"Long short-term memory (LSTM) is an artificial neural network used in the fields of artificial intelligence and deep learning. 

Unlike standard feedforward neural networks, LSTM has feedback connections. 

Such a recurrent neural network (RNN) can process not only single data points (such as images), but also entire sequences of data (such as speech or video). 

For example, LSTM is applicable to tasks such as unsegmented, connected:

* handwriting recognition,

* speech recognition,

* machine translation,

* robot control,

* video games,

* and healthcare.

LSTM has become the most cited neural network of the 20th century." !!! 

------

GRU

Source:

https://en.m.wikipedia.org/wiki/Gated_recurrent_unit

"The GRU is like a long short-term memory (LSTM) with a forget gate, but has fewer parameters than LSTM, as it lacks an output gate. 

GRU's performance on certain tasks of polyphonic music modeling, speech signal modeling and natural language processing was found to be similar to that of LSTM. 

GRUs have been shown to exhibit better performance on certain smaller and less frequent datasets."