Scheduler
|
#include <qLearning.h>
Public Member Functions | |
QLearning (std::shared_ptr< Context > context, TabularActionValues *actionValues) | |
void | init () override |
void | updateActionValues (state_t previousState, state_t nextState, action_t previousAction, double reward) override |
void | end () override |
![]() | |
RlBackupAlgorithm (std::shared_ptr< Context > c, ActionValuesFunction *av) | |
virtual | ~RlBackupAlgorithm () |
virtual double | getMaxQ (state_t state) |
virtual std::pair< action_t, double > | getBestActionAndQ (state_t state) |
virtual action_t | getBestAction (state_t state) |
virtual void | updateBestActionAndQ (state_t state) |
virtual void | notifyUpdateNeeded () |
Static Public Attributes | |
static constexpr const char * | configKey = "QLearning" |
Additional Inherited Members | |
![]() | |
virtual void | initAlpha () |
virtual void | updateAlpha () |
virtual void | updateIfNeeded (state_t state) |
![]() | |
std::shared_ptr< Context > | context {nullptr} |
ActionValuesFunction * | actionValues {nullptr} |
double | alpha {-1.0} |
double | alpha0 {0.1} |
double | alphaCounter {1.0} |
double | alphaDecaySpeed {1.0} |
bool | hyperbolic {false} |
bool | stepwise {false} |
unsigned long long | stepwiseCounter {0} |
unsigned long long int | stepLength {0} |
std::vector< double > | bestQ |
std::vector< action_t > | bestAction |
std::vector< bool > | needsUpdate |
Definition at line 23 of file qLearning.h.
QLearning::QLearning | ( | std::shared_ptr< Context > | context, |
TabularActionValues * | actionValues | ||
) |
Definition at line 23 of file qLearning.cpp.
|
overridevirtual |
Implements Mdp::RlBackupAlgorithm.
Definition at line 40 of file qLearning.cpp.
|
overridevirtual |
Implements Mdp::RlBackupAlgorithm.
Definition at line 29 of file qLearning.cpp.
|
overridevirtual |
Implements Mdp::RlBackupAlgorithm.
Definition at line 46 of file qLearning.cpp.
|
static |
Definition at line 26 of file qLearning.h.