mutUniformResetSHW

mutUniformMetaResetSHW

<code>[integer]</code> binary individual.

<code>[numeric]</code> average reset probability, must be between 0 and 1.

<code>[numeric]</code> approximate probability to draw 1-bit per entry.

reset.dist

further arguments passed on to the method.

<code>[matrix]</code> columns of probabilities, with <code>length(ind)</code> cols and
<code>length(reset.dist.weights)</code> rows.

reset.dists

<code>[numeric]</code> weight vector to select among <code>reset.dist</code> columns.

reset.dist.weights

Combination of the idea of <code>mutBitflipCHW</code> with <code>mutUniformReset</code>.
If a given bitvector has <code>m</code> 1s and <code>n</code> 0s, then, with probability <code>p</code> for
each bit, it is drawn anew from the distribution
<code>((m + 1) * reset.dist) / (m * reset.dist + n * (1 - reset.dist) + 1)</code>.
The reasoning behind this is that, without Laplace smoothing, drawing from
<code>m * reset.dist</code> / (m * reset.dist + n * (1 - reset.dist))<code>would lead to probabilities of drawing a "0" or "1" such that</code>mean(P("1") / P("0")) = m / n * mean(reset.dist / (1 - reset.dist))`.
The <code>mutUniformMetaResetSHW</code> does reset with a weighted mean of distributions.

Performs simultaneous hyperparameter tuning and
feature selection through both single-objective and multi-objective
optimization as described in Binder, Moosbauer et al. (2019) <arXiv:1912.12912>.
Uses the 'ecr'-package as basis but adds mixed integer
evolutionary strategies and multi-fidelity functionality as well as operators
specific for the problem of feature selection.

Martin Binder

mosmafs

Multi-Objective Simultaneous Model and Feature Selection

Susanne Dandl

Julia Moosbauer

mutUniformResetSHW function

<dl><dt>ind</dt>
<dd><code>[integer]</code> binary individual.</dd>
<dt>p</dt>
<dd><code>[numeric]</code> average reset probability, must be between 0 and 1.</dd>
<dt>reset.dist</dt>
<dd><code>[numeric]</code> approximate probability to draw 1-bit per entry.</dd>
<dt>...</dt>
<dd>further arguments passed on to the method.</dd>
<dt>reset.dists</dt>
<dd><code>[matrix]</code> columns of probabilities, with <code>length(ind)</code> cols and
<code>length(reset.dist.weights)</code> rows.</dd>
<dt>reset.dist.weights</dt>
<dd><code>[numeric]</code> weight vector to select among <code>reset.dist</code> columns.</dd></dl>

Arguments

Uniform Reset Scaled by Hamming Weight — mutUniformResetSHW

<dl>

<dt>ind</dt>
<dd><code>[integer]</code> binary individual.</dd>


<dt>p</dt>
<dd><code>[numeric]</code> average reset probability, must be between 0 and 1.</dd>


<dt>reset.dist</dt>
<dd><code>[numeric]</code> approximate probability to draw 1-bit per entry.</dd>


<dt>...</dt>
<dd>further arguments passed on to the method.</dd>


<dt>reset.dists</dt>
<dd><code>[matrix]</code> columns of probabilities, with <code>length(ind)</code> cols and
<code>length(reset.dist.weights)</code> rows.</dd>


<dt>reset.dist.weights</dt>
<dd><code>[numeric]</code> weight vector to select among <code>reset.dist</code> columns.</dd>

</dl>

mutUniformResetSHW: Uniform Reset Scaled by Hamming Weight

Description

Usage

Value

Arguments