condDist

condDist2

Finds a low-dimensional embedding of high-dimensional data, conditioning on available manifold information. The current version supports conditional MDS (based on either conditional SMACOF in Bui (2021) <arXiv:2111.13646> or closed-form solution in Bui (2022) <doi:10.1016/j.patrec.2022.11.007>) and conditional ISOMAP in Bui (2021) <arXiv:2111.13646>.

Anh Bui

Conditional Manifold Learning

condDist function

<dl> <dt>U</dt>
<dd>the embedding <code>U</code></dd> <dt>V.tilda</dt>
<dd><code>= V %*% B</code></dd> <dt>V.tilda2</dt>
<dd><code>= V %*% b^2*t(V)</code></dd> <dt>one_n_t</dt>
<dd><code>= t(rep(1,nrow(U)))</code></dd></dl>

Arguments

Author

Conditional Euclidean distance — condDist

<dl>

 <dt>U</dt>
<dd>the embedding <code>U</code></dd>

 <dt>V.tilda</dt>
<dd><code>= V %*% B</code></dd>

 <dt>V.tilda2</dt>
<dd><code>= V %*% b^2*t(V)</code></dd>

 <dt>one_n_t</dt>
<dd><code>= t(rep(1,nrow(U)))</code></dd>

</dl>

condDist: Conditional Euclidean distance

Description

Usage

Value

Arguments

Author

References