Time Series Forecasting with Deep Learning

The Encoder-Decoder Framework

Many of the models for [[time series forecasting]] The Time Series Forecasting Problem Forecasting time series using deep learning are following some sort of encoder-decoder architecture.

  1. Encoder: $g_{\text{enc}}(x^{(i)} _ {t-K:t}, u^{(i)} _ {t-K:t}) \to z_t$,
  2. Decoder: $g_{\text{dec}}(z_t, u^{(i)} _ {t+1: t+H}) \to y_{t+1:t+H}$.

Planted: by ;

No backlinks identified. Reference this note using the Note ID wiki/forecasting/forecasting-with-deep-learning.md in other notes to connect them.

L Ma (2022). 'Time Series Forecasting with Deep Learning', Datumorphism, 04 April. Available at: https://datumorphism.leima.is/wiki/forecasting/forecasting-with-deep-learning/.