Gated Activation Unit ã¯æ´»æ§åé¢æ°/ã¦ãããã®ä¸ç¨®.
output = tanh(Wfilter â input) ⦿ Ï(Wgate â input)
tanh(conv(input))ã§éç·å½¢å¤æããåºåã«å¯¾ããsigmoid(conv'(input)) ã§åºã¦ãã 0~1 ãç¨ããGatingãããã¦ããã¨ã¿ãªãã.
Gated PixelRNNã«ããã¦ãLSTMã®ã²ã¼ãããçæ³ãå¾ã¦éçº ( Gated Convolutional Layer).
WaveNetã§ã¯Gated Activation Unitã¨å¼ç§°ããã¦ãã.
ç¾ä»£çã«è¦ãã°attentionã«è¿ãããã®ãæãã.
Attentionã ã¨å
¨åãè¦ããããã£ã¡ã¯Gating weightè¨ç®ãConvã§ããã®ã§å±æãè¦ã¦Attentionã決ãã¦ã (MLP Attentionã«ããããªæãã®ãã£ãæ°ããã. ãããåé¢é£?).
é¢é£ããå è¡ç 究ã¯highway networks, grid LSTM (arXiv:1507.01526), neural GPUs (arXiv:1511.08228) ãªã©