神經(jīng)網(wǎng)絡中的權(quán)重衰減（L2正則化）與Dropout相比，主要區(qū)別是：

神經(jīng)網(wǎng)絡中的權(quán)重衰減（L2正則化）與Dropout相比，主要區(qū)別是：
A. 權(quán)重衰減通過懲罰大權(quán)重來防止過擬合，而Dropout通過隨機丟棄神經(jīng)元來防止過擬合
B. Dropout通過懲罰大權(quán)重來防止過擬合，而權(quán)重衰減通過隨機丟棄神經(jīng)元來防止過擬合
C. 權(quán)重衰減和Dropout都是通過懲罰大權(quán)重來防止過擬合
D. 權(quán)重衰減和Dropout都是通過隨機丟棄神經(jīng)元來防止過擬合