【讀書1】【2017】MATLAB與深度學習——過度擬合(1)

該方法儘可能簡化神經網絡的結構，從而減少過擬合的發生。（編注：正則化的簡化實質也是丟棄網絡節點，不過正則化是按照某種規則丟棄，而dropout是隨機丟棄） This method works as it simplifies theneural network’ architecture as much as possible, and hence reduces thepossible onse

>>阅读原文<<