2016AAAI_Face model compression by distilling knowledge from neurons (商湯)

論文地址    http://www.aaai.org/ocs/index.php/AAAI/AAAI16/paper/download/11977/12130 總結     代替Distilling the knowledge in a neural network (Hinton 2015)中作爲知識的soft target,從隱藏層的神經元學習知識,同時考慮到神經元可能包含噪聲或無關信息,需
相關文章
相關標籤/搜索