BP Algorithm ·正向过程：=wa-》=∑w4 a)f(n)) 输出层所有神经元的误差能量总和sum squared error,SSE J-se-2∑4，-a,) 。权值修正：delta学习规侧则 △w,=-n0w) onL=-n on aJ =-n an;owj a(k-1) 2023/7/9 102023/7/9 10 BP Algorithm ⚫ 正向过程: ( ) ( ) ( 1) ( ) ( 1) k k k k k j j ij i i n w a − − = = w a  ( ) ( ) ( ) k k a f n j j = ⚫ 输出层所有神经元的误差能量总和 sum squared error, SSE 1 2 sse ( ) 2 j j j J t a = = −  ⚫ 权值修正: delta学习规则 j ( 1) k j j j j j J J J n n n    −      = − = − = −     w a w w