Free cookie consent management tool by TermsFeed Policy Generator

Ignore:
Timestamp:
01/18/15 18:24:58 (8 years ago)
Author:
gkronber
Message:

#2283 fixed compile errors and refactoring

File:
1 edited

Legend:

Unmodified
Added
Removed
  • branches/HeuristicLab.Problems.GrammaticalOptimization/HeuristicLab.Algorithms.Bandits/IPolicy.cs

    r11770 r11793  
    1111  // here we assume that a reward is only recieved at the end of the episode and the update is done only after an episode is complete
    1212  // we also assume that the policy can fail to select one of the followStates
    13   public interface IPolicy<TState> {
    14     bool TrySelect(Random random, TState curState, IEnumerable<TState> afterStates, out TState selectedState); // selectedState \in afterStates
     13  public interface IPolicy<in TState> {
     14    bool TrySelect(Random random, TState curState, IEnumerable<TState> afterStates, out int selectedStateIdx); // selectedState \in afterStates
    1515
    1616    // state-trajectory are the states of the episode, at the end we recieved the reward (only for the terminal state)
Note: See TracChangeset for help on using the changeset viewer.