Computes the Freidlin-Wentzell quasi-potential between \(x_0\) and \(x_1\), constructs a tilted likelihood proportional to \(\exp(-V/\mathrm{eps})\), and fits a flow-based variational posterior.
fitflow_FW(
observed,
states = NULL,
flowtype = "maf",
flowspec = list(),
inittheta = NULL,
drift,
x0,
x1,
T = 200,
dt = 0.01,
eps = 0.1,
nmc = 256,
control = list()
)Output of fitflowvariational().
Empirical distribution Q.
Optional category names.
Flow type.
Structural parameters for the flow.
Optional initial theta.
Drift function \(b(x)\).
Starting point.
Target point.
Number of time steps.
Time step.
Noise strength (small parameter).
Monte Carlo samples.
Control list for optim().
This is useful for rare-event inference in small-noise diffusions, where the quasi-potential acts as an effective energy landscape.