矩阵求导

矩阵对标量求导:
U=U(x,y)U = U(x,y)U=U(x,y)
U 是矩阵, x,y 是标量.
∂∣U∣∂x=∣U∣tr(U−1∂U∂x)\frac{\partial |U|}{\partial x}=|U|tr(U^{-1}{\frac{\partial U}{\partial x}})xU=Utr(U1xU)

∂U−1∂x=−U−1∂U∂xU−1\frac{\partial U^{-1}}{\partial x}=-U^{-1}{\frac{\partial U}{\partial x}}U^{-1}xU1=U1xUU1

∂ln∣U∣∂x=tr(U−1∂U∂x)\frac{\partial ln|U|}{\partial x}=tr(U^{-1}{\frac{\partial U}{\partial x}})xlnU=tr(U1xU)

∂2∣U∣∂x2=∣U∣(tr(U−1∂2U∂x2)+(tr(U−1∂U∂x))2−tr(U−1∂U∂xU−1∂U∂x)) \frac{\partial^2 |U|}{\partial x^2}=|U|(tr(U^{-1}{\frac{\partial^2 U}{\partial x^2}})+(tr(U^{-1}{\frac{\partial U}{\partial x}}))^2 - tr(U^{-1}{\frac{\partial U}{\partial x}}U^{-1}{\frac{\partial U}{\partial x}}))x22U=U(tr(U1x22U)+(tr(U1xU))2tr(U1xUU1xU))

∂2U∂x2=U−1(∂U∂xU−1∂U∂x−∂2U∂x2+∂U∂xU−1∂U∂x)U−1\frac{\partial^2 U}{\partial x^2}= U^{-1}({\frac{\partial U}{\partial x}}U^{-1}{\frac{\partial U}{\partial x}} - {\frac{\partial^2 U}{\partial x^2}} + {\frac{\partial U}{\partial x}}U^{-1}{\frac{\partial U}{\partial x}})U^{-1}x22U=U1(xUU1xUx22U+xUU1xU)U1

∂2U∂x∂y=U−1(∂U∂xU−1∂U∂y−∂2U∂x∂y+∂U∂yU−1∂U∂x)U−1\frac{\partial^2 U}{\partial x\partial y}= U^{-1}({\frac{\partial U}{\partial x}}U^{-1}{\frac{\partial U}{\partial y}} - {\frac{\partial^2 U}{\partial x \partial y}} + {\frac{\partial U}{\partial y}}U^{-1}{\frac{\partial U}{\partial x}})U^{-1}xy2U=U1(xUU1yUxy2U+yUU1xU)U1

评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值