actionDimensions | rl::WireFitQLearn | |
adjustExploration(double uncertainty) | rl::FidoControlSystem | protected |
applyReinforcementToLastAction(double reward, State newState) | rl::FidoControlSystem | virtual |
baseOfDimensions | rl::WireFitQLearn | |
bestAction(State state) | rl::WireFitQLearn | protected |
chooseBestAction(State currentState) | rl::WireFitQLearn | virtual |
chooseBoltzmanAction(State currentState, double explorationConstant) | rl::WireFitQLearn | virtual |
chooseBoltzmanActionDynamic(State state) | rl::FidoControlSystem | |
controlPointsGDErrorTarget | rl::WireFitQLearn | |
controlPointsGDLearningRate | rl::WireFitQLearn | |
controlPointsGDMaxIterations | rl::WireFitQLearn | |
devaluationFactor | rl::WireFitQLearn | |
explorationLevel | rl::FidoControlSystem | |
FidoControlSystem(int stateDimensions, Action minAction, Action maxAction, int baseOfDimensions) | rl::FidoControlSystem | |
getError(std::vector< double > input, std::vector< double > correctOutput) | rl::FidoControlSystem | protected |
getQValue(double reward, const State &oldState, const State &newState, const Action &action, const std::vector< Wire > &controlWires) | rl::WireFitQLearn | protected |
getRawOutput(std::vector< Wire > wires) | rl::WireFitQLearn | protected |
getSetOfWires(const State &state, int baseOfDimensions) | rl::WireFitQLearn | protected |
getWires(State state) | rl::WireFitQLearn | protected |
highestReward(State state) | rl::WireFitQLearn | protected |
histories | rl::FidoControlSystem | protected |
initialExploration | rl::FidoControlSystem | |
interpolator | rl::WireFitQLearn | |
lastAction | rl::WireFitQLearn | |
lastState | rl::WireFitQLearn | |
lastUncertainty | rl::FidoControlSystem | |
learningRate | rl::WireFitQLearn | |
maxAction | rl::WireFitQLearn | |
minAction | rl::WireFitQLearn | |
modelNet | rl::WireFitQLearn | |
network | rl::WireFitQLearn | |
newControlWires(const Wire &correctWire, std::vector< Wire > controlWires) | rl::WireFitQLearn | protected |
newControlWiresForHistory(History history) | rl::FidoControlSystem | protected |
numberOfWires | rl::WireFitQLearn | |
reset() | rl::FidoControlSystem | virtual |
samplesOfHistory | rl::FidoControlSystem | |
selectHistories() | rl::FidoControlSystem | protected |
store(std::ofstream *output) | rl::WireFitQLearn | |
trainer | rl::WireFitQLearn | |
trainOnHistories(std::vector< FidoControlSystem::History > selectedHistories) | rl::FidoControlSystem | protected |
WireFitQLearn(unsigned int stateDimensions, unsigned int actionDimensions_, unsigned int numHiddenLayers, unsigned int numNeuronsPerHiddenLayer, unsigned int numberOfWires_, Action minAction_, Action maxAction_, unsigned int baseOfDimensions_, Interpolator *interpolator_, net::Trainer *trainer_, double learningRate_, double devaluationFactor_) | rl::WireFitQLearn | |
WireFitQLearn() | rl::WireFitQLearn | |
WireFitQLearn(std::ifstream *input) | rl::WireFitQLearn | |