rurel

use rurel::mdp::State;

#[derive(PartialEq, Eq, Hash, Clone)]
struct MyState { x: i32, y: i32 }
#[derive(PartialEq, Eq, Hash, Clone)]
struct MyAction { dx: i32, dy: i32 }

impl State for MyState {
	type A = MyAction;
	fn reward(&self) -> f64 {
		// Negative Euclidean distance
		-((((10 - self.x).pow(2) + (10 - self.y).pow(2)) as f64).sqrt())
	}
	fn actions(&self) -> Vec<MyAction> {
		vec![MyAction { dx: 0, dy: -1 },	// up
			 MyAction { dx: 0, dy: 1 },	// down
			 MyAction { dx: -1, dy: 0 },	// left
			 MyAction { dx: 1, dy: 0 },	// right
		]
	}
}

Then define the agent:

use rurel::mdp::Agent;

struct MyAgent { state: MyState }
impl Agent<MyState> for MyAgent {
	fn current_state(&self) -> &MyState {
		&self.state
	}
	fn take_action(&mut self, action: &MyAction) -> () {
		match action {
			&MyAction { dx, dy } => {
				self.state = MyState {
					x: (((self.state.x + dx) % 21) + 21) % 21, // (x+dx) mod 21
					y: (((self.state.y + dy) % 21) + 21) % 21, // (y+dy) mod 21
				}
			}
		}
	}
}

That's all. Now make a trainer and train the agent with Q learning, with learning rate 0.2, discount factor 0.01 and an initial value of Q of 2.0. We let the trainer run for 100000 iterations, randomly exploring new states.

use rurel::AgentTrainer;
use rurel::strategy::learn::QLearning;
use rurel::strategy::explore::RandomExploration;
use rurel::strategy::terminate::FixedIterations;

let mut trainer = AgentTrainer::new();
let mut agent = MyAgent { state: MyState { x: 0, y: 0 }};
trainer.train(&mut agent,
              &QLearning::new(0.2, 0.01, 2.),
              &mut FixedIterations::new(100000),
              &RandomExploration::new());

After this, you can query the learned value (Q) for a certain action in a certain state by:

trainer.expected_value(&state, &action) // : Option<f64>

Development

• Run cargo fmt --all to format the code.
• Run cargo clippy --all-targets --all-features -- -Dwarnings to lint the code.
• Run cargo test to test the code.