Struct EpsilonGreedy 

pub struct EpsilonGreedy { /* private fields */ }

Epsilon-greedy multi-armed bandit.

With probability epsilon, selects a uniformly random arm (exploration). Otherwise, selects the arm with the highest estimated value (exploitation), breaking ties by lowest index.

When constructed with EpsilonGreedy::with_decay, epsilon decays exponentially as initial_epsilon * decay^t after each update, where t is the total number of pulls so far.

Example

use irithyll::bandits::{Bandit, EpsilonGreedy};

let mut bandit = EpsilonGreedy::new(3, 0.1);
for _ in 0..100 {
    let arm = bandit.select_arm();
    let reward = if arm == 0 { 1.0 } else { 0.0 };
    bandit.update(arm, reward);
}
assert!(bandit.arm_values()[0] > bandit.arm_values()[1]);

Implementations

impl EpsilonGreedy

pub fn new(n_arms: usize, epsilon: f64) -> Self

Create an epsilon-greedy bandit with fixed epsilon and default seed (42).

pub fn with_seed(n_arms: usize, epsilon: f64, seed: u64) -> Self

Create an epsilon-greedy bandit with a custom RNG seed.

pub fn with_decay(n_arms: usize, epsilon: f64, decay: f64) -> Self

Create an epsilon-greedy bandit with exponential epsilon decay.

After each update, epsilon becomes initial_epsilon * decay^t where t is the total number of pulls. A typical decay value is 0.999.

pub fn current_epsilon(&self) -> f64

Current epsilon value (may differ from initial if decay is active).

Trait Implementations

impl Bandit for EpsilonGreedy

fn select_arm(&mut self) -> usize

Select the next arm to pull.

fn update(&mut self, arm: usize, reward: f64)

Update the bandit with observed reward for the given arm.

fn n_arms(&self) -> usize

Number of arms.

fn n_pulls(&self) -> u64

Total number of pulls across all arms.

fn reset(&mut self)

Reset to initial state (all counts and estimates zeroed).

fn arm_values(&self) -> &[f64]

Per-arm estimated reward values.

fn arm_counts(&self) -> &[u64]

Per-arm pull counts.

impl Clone for EpsilonGreedy

fn clone(&self) -> EpsilonGreedy

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for EpsilonGreedy

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

Formats the value using the given formatter.