LSTM

Long Short Term Memory (LSTM)

LSTM is a type of recurrent neural network (RNN) architecture designed to address the vanishing gradient problem that can occur in traditional RNNs.
It introduces memory cells and various gates to control the flow of information through the network. The key components of an LSTM unit are the input gate, forget gate, cell state, and output gate .

LSTM Components

Let's break down the LSTM architecture with a simple example:

1. Input Gate (i):
   Responsible for deciding which information from the input should be stored in the cell state. It takes the current input and the previous hidden state as input and produces a candidate new cell state.
2. Forget Gate (f):
   Determines what information from the previous cell state should be discarded or kept. Takes the current input and the previous hidden state as input and produces a forget factor for each element in the cell state.
3. Cell State (c):
   This is the memory of the cell. Updated by combining the input from the input gate and the forget gate.
4. Output Gate (o):
   Decides what the next hidden state should be. Takes the current input and the previous hidden state, along with the updated cell state, and produces the next hidden state.

Detailed Explanation

Disclaimer

Creating this blog only for my revision purpose and not for any monetary gain.