Package jline.lang.reward

Interface RewardFunction

All Superinterfaces:
Serializable
Functional Interface:
This is a functional interface and can therefore be used as the assignment target for a lambda expression or method reference.
@FunctionalInterface public interface RewardFunction extends Serializable
Functional interface for defining reward functions on CTMC states. A reward function maps a state vector and network structure to a scalar reward value. This interface is used for CTMC reward computation via value iteration. Example usage: 
 // Queue length reward
 RewardFunction qlenReward = (state, sn) -> state.get(0, 1);

 // Throughput reward
 RewardFunction tputReward = (state, sn) -> {
     double n = state.get(0, 1);
     return n > 0 ? n * sn.rates.get(1, 0) / n : 0;
 };

  • Method Summary

    Modifier and Type
    Method
    Description
    double
    compute(Matrix state, NetworkStruct sn)
    Compute the reward value for a given state.

  • Method Details

    • compute

      double compute(Matrix state, NetworkStruct sn)
      Compute the reward value for a given state.
      Parameters:
      state - The state vector (row from stateSpaceAggr matrix) Format: [n_{1,1}, n_{1,2}, ..., n_{M,K}] where n_{i,k} is the number of class-k jobs at station i
      sn - The NetworkStruct providing access to rates, parameters, etc.
      Returns:
      The reward value for this state