逻辑回归问题中的代价函数为:
J = − 1 m ∑ i = 1 m [ y ( i ) l o g h θ ( x ( i ) ) + ( 1 − y ( i ) ) l o g ( 1 − h θ ( x ( i ) ) ) ] J=-\frac{1}{m}\sum\limits_{i=1}^{m}{[y^{(i)}logh_\mathbf{\theta}(\mathbf{x}^{(i)})+(1-y^{(i)})log(1-h_\mathbf{\theta}(\mathbf{x}^{(i)}))]} J=−m1i=1∑m[y(i)loghθ(x(i))+(1−y(i))log(1−hθ(x(i)))],其中, h θ ( x ) = g ( θ T x ) , g ( z ) = 1 1 + e − z h_\theta(x)=g(\mathbf{\theta}^T\mathbf{x}), g(z)=\frac{1}{1+e^{-z}} hθ(x)=g(θTx), g(z)=1+e−z1。
其关于参数 θ \mathbf{\theta} θ的偏导数为:
∂ J ∂ θ = 1 m ∑ i = 1 m [ h θ ( x ( i ) ) − y ( i ) ] x ( i ) \frac{\partial{J}}{\partial{\mathbf{\theta}}}=\frac{1}{m}\sum\limits_{i=1}^{m}{[h_{\mathbf{\theta}}(\mathbf{x}^{(i)})-y^{(i)}]\mathbf{x}^{(i)}} ∂θ∂J=m1i=1∑m[hθ(x(i))−y(i)]x(i)
下面介绍一下偏导数推导过程:
偏导数推导的原理是链式法则,即:若有 y = f ( g ( x ) ) y=f(g(x)) y=f(g(x)),则有 ∂ y ∂ x = ∂ f ∂ g ⋅ ∂ g ∂ x \frac{\partial{y}}{\partial{x}}=\frac{\partial{f}}{\partial{g}}\cdot{\frac{\partial{g}}{\partial{x}}} ∂x∂y=∂g∂f⋅∂x∂g。
运用链式法则,有
∂ J ∂ θ = ∂ J ∂ h ∂ h ∂ g ∂ g ∂ θ \frac{\partial{J}}{\partial{\mathbf{\theta}}}=\frac{\partial{J}}{\partial{h}}\frac{\partial{h}}{\partial{g}}\frac{\partial{g}}{\partial{\mathbf{\theta}}} ∂θ∂J=∂h∂J∂g
逻辑回归中偏导数的推导详解
最新推荐文章于 2023-01-17 18:17:27 发布