AI Blog

AI Blog

by Michele Laurelli

RNN (Recurrent Neural Network)

/rɪˈkʌrənt ˈnjʊərəl ˈnetwɜːrk/
Architecture
Definition

A neural network architecture designed for sequential data, with connections that loop back to previous states.

RNNs process sequences by maintaining a hidden state that captures information from previous time steps. They excel at tasks with temporal dependencies but suffer from vanishing gradients. LSTM and GRU address this issue.

Examples

1

Time series forecasting

2

Text generation

3

Speech recognition

Michele Laurelli - AI Research & Engineering