Cross-entropy method: Difference between revisions

Revision as of 18:18, 14 July 2016

Revision as of 21:16, 14 July 2016

Line 22:

# Choose initial parameter vector <math>\mathbf{v}^{(0)}</math>; set t = 1.

# Generate a random sample <math>\mathbf{X}_1,\dots,\mathbf{X}_N</math> from <math>f(\cdot;\mathbf{v}^{(t-1)})</math>

# Solve for <math>\mathbf{v}^{(t)}</math>, where<br><math>\mathbf{v}^{(t)} = \mathop{\textrm{argmax}}_{\mathbf{vu}} \frac{1}{N} \sum_{i=1}^N H(\mathbf{X}_i)\frac{f(\mathbf{X}_i;\mathbf{u})}{f(\mathbf{X}_i;\mathbf{v}^{(t-1)})} \log f(\mathbf{X}_i;\mathbf{v}^{(t-1)})</math>

# If convergence is reached then '''stop'''; otherwise, increase t by 1 and reiterate from step 2.