發(fā)簡(jiǎn)信
IP屬地：湖北
-
《Neural Baby Talk》 CVPR2018與 《Knowing When to Look: Adaptive Attention via A Visual Sentinel for Im...
-
Independently Recurrent Neural Network (IndRNN): Building A Longer and Deeper RNNMotivation plain的RNN存在梯度消失與爆炸問題，而LSTM/GRU雖然一定程度解決了問題，但是由于采用tanh或者sigmoid...