Function mnl_select 

pub fn mnl_select<R: Rng>(costs: &[f64], theta: f64, rng: &mut R) -> usize

Select one alternative from a set of costs using the multinomial logit model.

Arguments

• costs - generalized cost of each alternative (lower = better)
• theta - scale parameter controlling dispersion. Larger θ means more deterministic (strongly favors lowest cost). Smaller θ means more random.
• rng - random number generator

Returns

Index of the selected alternative in the costs slice.

Edge cases

• If costs is empty, returns 0.
• If costs has one element, returns 0.
• If all utilities are degenerate (overflow/underflow), returns 0.