Enum AreaScalarisation 

pub enum AreaScalarisation {
    ExpectedValue,
    WorstCase {
        inner_maxiter: usize,
        inner_seed: u64,
    },
    Cvar {
        alpha: f64,
    },
}

How to scalarise the Q per-point losses into one outer-loop loss.

Variants

ExpectedValue

Probability-weighted mean: ∫ L(x,p) π(p) dp. The standard “expected loss over the listening area”.

WorstCase

max_{p ∈ support(π)} L(x, p). Robust / minimax. Implemented by inner-search over the bounding box (ignores the density shape; the max is taken over the support).

Fields

inner_maxiter: usize

Inner-search budget. 50 is usually plenty for D ≤ 3.

inner_seed: u64

Inner-search seed.

Cvar

Conditional Value-at-Risk at level α: mean of the worst α-fraction of per-point losses. alpha = 0.1 averages the worst 10 %; alpha = 1.0 degenerates to AreaScalarisation::ExpectedValue.

Fields

alpha: f64

Tail fraction in (0, 1].

Trait Implementations

impl Clone for AreaScalarisation

fn clone(&self) -> AreaScalarisation

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for AreaScalarisation

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Copy for AreaScalarisation