Module advanced 

Advanced multi-objective optimization utilities (clean Vec<f64>-based API).

Provides:

• NsgaIII — struct-based NSGA-III many-objective optimizer
• NsgaResult — unified result type
• Free functions: dominates, non_dominated_sort, hypervolume, crowding_distance, weighted_sum_optimize, epsilon_constraint_optimize

All functions operate on Vec<f64> and &[f64] rather than ndarray types, making them easy to use from pure Rust code without array wrappers.

Example

use scirs2_optimize::multi_objective::advanced::{NsgaIII, dominates, non_dominated_sort};

assert!(dominates(&[1.0, 2.0], &[2.0, 3.0]));

let sols = vec![vec![1.0, 3.0], vec![2.0, 2.0], vec![3.0, 1.0], vec![2.0, 3.0]];
let fronts = non_dominated_sort(&sols);
assert_eq!(fronts[0].len(), 3);

Structs

NsgaIII

NSGA-III many-objective optimizer — struct-based API.

NsgaResult

Result from NsgaIII::optimize.

Functions

crowding_distance

Compute crowding distances for a single Pareto front.

dominates

Returns true if objective vector a Pareto-dominates b.

epsilon_constraint_optimize

Epsilon-constraint method for multi-objective optimization.

hypervolume

Compute the hypervolume indicator.

non_dominated_sort

Fast non-dominated sorting (NSGA-II style).

weighted_sum_optimize

Minimize a weighted sum of objectives via multi-start coordinate descent.