Enum AgentPolicy 

pub enum AgentPolicy {
    EpsilonGreedy {
        epsilon: f64,
        decay: f64,
        min_epsilon: f64,
    },
    Boltzmann {
        temperature: f64,
    },
    UCB {
        c: f64,
    },
    Random,
}

Action-selection policy used by the agent.

Variants

EpsilonGreedy

With probability epsilon pick a random action; otherwise pick argmax Q(s,·).

Fields

epsilon: f64

Current exploration probability ∈ [0, 1].

decay: f64

Multiplicative decay applied each call to decay_epsilon.

min_epsilon: f64

Lower bound for epsilon after decay.

Boltzmann

Softmax action selection with temperature τ > 0.

Fields

temperature: f64

Temperature τ — higher → more uniform; lower → more greedy.

UCB

Upper-Confidence-Bound action selection.

Fields

c: f64

Exploration coefficient c ≥ 0.

Random

Uniform random action selection (pure exploration).

Trait Implementations

impl Clone for AgentPolicy

fn clone(&self) -> AgentPolicy

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AgentPolicy

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

Formats the value using the given formatter.

impl PartialEq for AgentPolicy

fn eq(&self, other: &AgentPolicy) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for AgentPolicy