Nonnegative matrix factorization

From Wikimization

(Difference between revisions)
Jump to: navigation, search
Current revision (17:40, 30 October 2016) (edit) (undo)
 
Line 9: Line 9:
by solving
by solving
-
<math>\begin{array}{cl}\mbox{find}_{A\in\mathbb{S}^3,\,B\in\mathbb{S}^3,\,W\in\mathbb{R}^{3\times2},\,H\in\mathbb{R}^{2\times3}}&W\,,\,H\\
+
<math>\begin{array}{cl}{\text find}_{A\in\mathbb{S}^3,\,B\in\mathbb{S}^3,\,W\in\mathbb{R}^{3\times2},\,H\in\mathbb{R}^{2\times3}}&W\,,\,H\\
-
\mbox{subject to}&Z=\left[\begin{array}{ccc}I&W^{\rm T}&H\\W&A&X\\H^{\rm T}&X^{\rm T}&B\end{array}\right]\succeq0\\
+
{\text subject to}&Z=\left[\begin{array}{ccc}I&W^{\rm T}&H\\W&A&X \\H^{\rm T}&X^{\rm T}&B\end{array}\right]\succeq0\\
&W\geq0\\
&W\geq0\\
&H\geq0\\
&H\geq0\\
-
&\mbox{rank}\,Z\leq2\end{array}</math>
+
&{\text rank}\,Z\leq2\end{array}</math>
which follows from the fact, at optimality,
which follows from the fact, at optimality,
Line 30: Line 30:
<math>\begin{array}{cl}\mbox{minimize}_{A\in\mathbb{S}^3,\,B\in\mathbb{S}^3,\,W\in\mathbb{R}^{3\times2},\,H\in\mathbb{R}^{2\times3}}&\langle Z\,,Y\rangle\\
<math>\begin{array}{cl}\mbox{minimize}_{A\in\mathbb{S}^3,\,B\in\mathbb{S}^3,\,W\in\mathbb{R}^{3\times2},\,H\in\mathbb{R}^{2\times3}}&\langle Z\,,Y\rangle\\
-
\mbox{subject to}&Z=\left[\begin{array}{ccc}I&W^{\rm T}&H\\W&A&X\\H^{\rm T}&X^{\rm T}&B\end{array}\right]\succeq0\\
+
\mbox{subject to}&Z=\left[\begin{array}{ccc}I&W^{\rm T}&H\\W&A&X \\H^{\rm T}&X^{\rm T}&B\end{array}\right]\succeq0\\
&W\geq0\\
&W\geq0\\
&H\geq0\end{array}</math>
&H\geq0\end{array}</math>

Current revision

Exercise from Convex Optimization & Euclidean Distance Geometry, ch.4:

Given rank-2 nonnegative matrix LaTeX: X=\!\left[\!\begin{array}{ccc}17&28&42\\
</p>
<pre>                                      16&47&51\\
</pre>
<p>17&82&72\end{array}\!\right], find a nonnegative factorization LaTeX:  X=WH\, by solving

LaTeX: \begin{array}{cl}{\text find}_{A\in\mathbb{S}^3,\,B\in\mathbb{S}^3,\,W\in\mathbb{R}^{3\times2},\,H\in\mathbb{R}^{2\times3}}&W\,,\,H\\
{\text subject to}&Z=\left[\begin{array}{ccc}I&W^{\rm T}&H\\W&A&X \\H^{\rm T}&X^{\rm T}&B\end{array}\right]\succeq0\\
&W\geq0\\
&H\geq0\\
&{\text rank}\,Z\leq2\end{array}

which follows from the fact, at optimality,

LaTeX:  Z^*=\left[\!\begin{array}{c}I\\W\\H^{\rm T}\end{array}\!\right]\begin{array}{c}\textbf{[}\,I~~W^{\rm T}~H\,\textbf{]}
\end{array}

Use the known closed-form solution for a direction vector LaTeX: Y\, to regulate rank (rank constraint is replaced) by Convex Iteration;

set LaTeX: _{}Z^*\!=Q\Lambda Q^{\rm T}\!\in\mathbb{S}^{\mathbf{8}} to a nonincreasingly ordered diagonalization and LaTeX: _{}U^*\!=_{\!}Q(:\,,_{^{}}3:8)\!\in_{\!}\mathbb{R}^{\mathbf{8}\times\mathbf{6}}, then LaTeX: Y\!=U^* U^{*\rm T}.


In summary, initialize LaTeX: Y=I\, then alternate solution of

LaTeX: \begin{array}{cl}\mbox{minimize}_{A\in\mathbb{S}^3,\,B\in\mathbb{S}^3,\,W\in\mathbb{R}^{3\times2},\,H\in\mathbb{R}^{2\times3}}&\langle Z\,,Y\rangle\\
\mbox{subject to}&Z=\left[\begin{array}{ccc}I&W^{\rm T}&H\\W&A&X \\H^{\rm T}&X^{\rm T}&B\end{array}\right]\succeq0\\
&W\geq0\\
&H\geq0\end{array}

with

LaTeX: Y\!=U^* U^{*\rm T}. Global convergence occurs, in this example, in only a few iterations.

Personal tools