next up previous contents
Next: B2. Hessian Matrix Up: B. Optimization Previous: B. Optimization


B1. Jacobian Matrix

The Jacobian matrix of a vector function $\vec{f}(\vec{x})= (f_{1}(\vec{x}), \dots
,f_{m}(\vec{x}))^{\cal T}$ ( $\vec{x} \in \mathbb{R}^{n}$) is defined by the matrix $\mathcal{J} \in \mathbb{R}^{m \times n}$


\begin{displaymath}
\mathcal{J}(\vec{x}) = \mathop{\rm Jac}\nolimits (\vec{f}(\...
...tial f_{m}(\vec{x})}{\partial x_n} \\
\end{array} \right )
.
\end{displaymath} (B1)

The Jacobian can also be written by the transposed gradient vector $f_{i}(\vec{x})$ for each row


\begin{displaymath}
\mathop{\rm Jac}\nolimits (\vec{x}) =
\left (
\begin{ar...
...mits f_{m}(\vec{x}) \right )^{\cal T}
\end{array} \right )
.
\end{displaymath} (B2)




R. Plasun