Long Short-Term Memory (LSTM)
Era: Revival · 1990s–2010s
Sepp Hochreiter and Jürgen Schmidhuber proposed LSTM, a special type of RNN capable of learning long-term dependencies. LSTM achieved great success in speech recognition and language modeling.
Sepp Hochreiter & Jürgen Schmidhuber
Sources and support
Long Short-Term Memory
Neural Computation · Source type: Paper
The original 1997 architecture uses memory cells, input gates, and output gates.
Published: ·Accessed:
Learning to Forget: Continual Prediction with LSTM
Neural Computation / PubMed · Source type: Paper
The later paper introduces the forget gate.
Published: ·Accessed: