splinepwlinflowmodel

A readable implementation of a monotone piecewise-linear spline flow.
Each dimension is transformed independently using learned spline parameters.

Variational flow-based methods for modeling rare events using
Kullback–Leibler (KL) divergence, normalizing flows, Girsanov change of
measure, and Freidlin–Wentzell action functionals. The package provides
tools for rare-event inference, minimum-action paths, and quasi-potential
computation in stochastic dynamical systems. Methods are based on
Rezende and Mohamed (2015) <doi:10.48550/arXiv.1505.05770>,
Girsanov (1960) <doi:10.1137/1105027>,
and Freidlin and Wentzell (2012, ISBN:978-0387955477).

Pietro Piu

rareflow

Variational Flow-Based Inference for Rare Events and Large
Deviations

splinepwlinflowmodel function

<dl><dt>d</dt>
<dd>Dimension of the latent space.</dd>
<dt>K</dt>
<dd>Number of spline bins.</dd>
<dt>theta</dt>
<dd>Optional parameter vector. If NULL, random initialization.</dd></dl>

Arguments

Piecewise-Linear Spline Flow (Monotone) — splinepwlinflowmodel

<dl>

<dt>d</dt>
<dd>Dimension of the latent space.</dd>


<dt>K</dt>
<dd>Number of spline bins.</dd>


<dt>theta</dt>
<dd>Optional parameter vector. If NULL, random initialization.</dd>

</dl>

splinepwlinflowmodel: Piecewise-Linear Spline Flow (Monotone)

Description

Usage

Value

Arguments

Details