Struct QLearner 

pub struct QLearner { /* private fields */ }

A simple reinforcement learning framework that can be used to learn optimal policies for Markov decision processes using Q-learning. Q-learning is a model-free reinforcement learning algorithm that learns an optimal action-value function from experience by repeatedly updating estimates of the Q-value of state-action pairs.

Implementations

impl QLearner

pub fn update(&mut self, state: u64, action: u32, reward: f64)

Updates Q-value for a state-action pair based on received reward.

Arguments

• state - An integer representing the state.
• action - An integer representing the action.
• reward - A number representing the reward received for the action in the state.

pub fn get_best_action(&mut self, state: u64) -> i32

Returns the best action for a given state based on the current Q-values.

Arguments

• state - The current state.

Returns

• i32 - The action with the highest Q-value for the given state.

pub fn visit_matrix(&mut self, handler: Box<dyn FnMut(u64, u32, f64)>)

Visits all state-action pairs and calls the provided handler function for each pair.

Arguments

• handler - A function that is called for each state-action pair.

pub fn pack(hints: &Vec<i32>, values: &Vec<i32>) -> u64

Constructs a state from given hints and condition values.

Arguments

• hints - A vector of integers representing the byte length of provided values.
• values - The condition values as discrete values.

Returns

• i64 - The packed state value.

pub fn unpack(hints: &Vec<i32>, state: u64) -> Vec<i32>

Deconstructs a state from given hints to get condition values.

Arguments

• hints - A vector of integers representing the byte length of provided values.
• state - The state integer to unpack.

Returns

• Vec<i32> - The condition values as discrete values.

pub fn new(gamma: f64, alpha: f64, max_q: f64) -> QLearner

Creates a new QLearner object with optional parameters for gamma, alpha, and maxQ.

Arguments

• gamma - The discount factor for future rewards.
• alpha - The learning rate for updating Q-values.
• maxQ - The maximum Q-value. Defaults to 100.0.

Returns

• QLearner - The newly created QLearner object.

Trait Implementations

impl Clone for QLearner

fn clone(&self) -> QLearner

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Drop for QLearner

fn drop(&mut self)

Executes the destructor for this type.

impl IObject for QLearner

fn raw(&self) -> i64

fn obj(&self) -> &dyn IObject

fn as_any(&self) -> &dyn Any

fn get_id(&self) -> i32