# 深度学习权重参数初始化要点

Cost after iteration 0: 0.693148
Cost after iteration 100: 0.678011
Cost after iteration 200: 0.667600
Cost after iteration 300: 0.660422
Cost after iteration 400: 0.655458
Cost after iteration 500: 0.652013
Cost after iteration 600: 0.649616
Cost after iteration 700: 0.647942
Cost after iteration 800: 0.646770
Cost after iteration 900: 0.645947
Cost after iteration 1000: 0.645368
Cost after iteration 1100: 0.644961
Cost after iteration 1200: 0.644673
Cost after iteration 1300: 0.644469
Cost after iteration 1400: 0.644325
Cost after iteration 1500: 0.644223
Cost after iteration 1600: 0.644151
Cost after iteration 1700: 0.644100
Cost after iteration 1800: 0.644063
Cost after iteration 1900: 0.644037
Cost after iteration 2000: 0.644019

## 一种比较简单、有效的方法是：权重参数初始化从区间均匀随机取值。

1.符合均匀分布U（a,b）的随机变量数学期望和方差分别是——数学期望：E(X)=(a+b)/2，方差：D(X)=(b-a)²/12

2.如果随机变量X,Y是相互独立的，那么Var(X+Y) = Var(X)+Var(Y)，如果X,Y是相互独立的且均值为0，那么Var(X*Y) = Var(X)*Var(Y)

xi)是均值=0，标准差=1的，那么

-------------------------------------------------------------------------------------------------

Cost after iteration 0: 0.715732
Cost after iteration 100: 0.674738
Cost after iteration 200: 0.660337
Cost after iteration 300: 0.646289
Cost after iteration 400: 0.629813
Cost after iteration 500: 0.606006
Cost after iteration 600: 0.569004
Cost after iteration 700: 0.519797
Cost after iteration 800: 0.464157
Cost after iteration 900: 0.408420
Cost after iteration 1000: 0.373155
Cost after iteration 1100: 0.305724
Cost after iteration 1200: 0.268102
Cost after iteration 1300: 0.238725
Cost after iteration 1400: 0.206323
Cost after iteration 1500: 0.179439
Cost after iteration 1600: 0.157987
Cost after iteration 1700: 0.142404
Cost after iteration 1800: 0.128652
Cost after iteration 1900: 0.112443
Cost after iteration 2000: 0.085056
Cost after iteration 2100: 0.057584
Cost after iteration 2200: 0.044568
Cost after iteration 2300: 0.038083

Cost after iteration 2400: 0.034411

• 广告
• 抄袭
• 版权
• 政治
• 色情
• 无意义
• 其他

120