Scheduler
watkinsQLambda.h
Go to the documentation of this file.
1 
10 #ifndef WATKINS_Q_LAMBDA_H
11 #define WATKINS_Q_LAMBDA_H
12 
13 
14 #include "eligibilityTraceAlgo.h"
15 
16 namespace Mdp
17 {
18 
19 
21 {
22 public:
23  static const constexpr char *configKey = "watkinsQLambda";
24 public:
25  WatkinsQLambda(std::shared_ptr<Context> c, TabularActionValues *av)
26  : EligibilityTraceAlgo(c, av)
27  , tabularAv(av)
28  {};
29  virtual void updateActionValues(state_t previousState, state_t mextState, action_t previousAction, double reward);
30 private:
31  std::pair<double, action_t> watkinsBestQ(state_t state, action_t nextAction);
32  TabularActionValues *tabularAv{nullptr};
33 };
34 
35 }
36 
37 
38 #endif
virtual void updateActionValues(state_t previousState, state_t mextState, action_t previousAction, double reward)
static const constexpr char * configKey
size_t action_t
Definition: action_impl.h:18
WatkinsQLambda(std::shared_ptr< Context > c, TabularActionValues *av)
Definition: action.h:18
size_t state_t
Definition: state.h:19
Definition: reward.py:1