LSTM及GRU整理。

发布时间 2023-12-11 15:15:44作者: lotuslaw
  • LSTM

\[I_t=\sigma(X_tW_{xi}+H_{t-1}W_{hi}+b_i)\\ F_t=\sigma(X_tW_{xf}+H_{t-1}W_{hf}+b_f)\\ O_t=\sigma(X_tW_{xo}+H_{t-1}W_{ho}+b_o)\\ \bar{C_t}=tanh(X_tW_{xc}+H_{t-1}W_{hc}+b_c)\\ C_t=F_t\odot{C_{t-1}}+I_t\odot{\bar{C_t}}\\ H_t=O_t\odot{tanh(C_t)} \]

  • GRU

\[R_t=\sigma(X_tW_{xr}+H_{t-1}W_{hr}+b_r)\\ Z_t=\sigma(X_tW_{xz}+H_{t-1}W_{hz}+b_z)\\ \bar{H_{t}}=tanh(X_tW_{xh}+(R_t\odot{H_{t-1}})W_{hh}+b_h)\\ H_t=Z_t\odot{H_{t-1}}+(1-Z_t)\odot{\bar{H_t}} \]