RewardFunction

scirs2_optimize::reinforcement_learning

Trait RewardFunction 

Source 
pub trait RewardFunction {
    // Required method
    fn compute_reward(
        &self,
        prev_state: &OptimizationState,
        action: &OptimizationAction,
        new_state: &OptimizationState,
    ) -> f64;
}
Expand description

Reward function for optimization RL

Required Methods§

Source

fn compute_reward( &self, prev_state: &OptimizationState, action: &OptimizationAction, new_state: &OptimizationState, ) -> f64

Compute reward based on state transition

Implementors§