Overview  Package  Class  Tree  Index  Help 
 PREV   NEXT FRAMES    NO FRAMES    

Serialized Form

Package grsnc.binb

Class grsnc.binb.Rivest06 implements Serializable

Serialized Fields

rnd

java.util.Random rnd

m_StateRep

StateRepresentation m_StateRep
StateRepresentation converting state into real-valued vector.

m_ActorCount

int m_ActorCount
Number of actor neurons.

m_CriticCount

int m_CriticCount
Number of critic neurons.

m_Wa

double[][] m_Wa
Stimuli to actor weights.

m_Wc

double[][] m_Wc
Stimuli to critic weights.

m_Gamma

double m_Gamma
Discounting factor.

m_Lambda

double m_Lambda
Eligibility trace discount factor.

m_Etaa

double m_Etaa
Actor learning rate.

m_Etac

double m_Etac
Critic learning rate.

m_InitWeightFactor

double m_InitWeightFactor
Initialization weight factor.

Package lnsc

Class lnsc.AbstractFunctionalUnit implements Serializable

Serialized Fields

m_InputCount

int m_InputCount
Indicates the number of variables of the function. That is, the value returned by getInputCount(). Derived classes must fill this slot in their constructor.

m_OutputCount

int m_OutputCount
Indicates the number of values returned by the function. Tha is, the value returned by getOutputCount(). Derived classes must fill this slot in their constructor.

m_IsDifferentiable

boolean m_IsDifferentiable
Indicates whether or not the function is differentiable. That is, the value returned by isDifferentiable(). Derived classes must fill this slot in their constructor.

m_IsTwiceDifferentiable

boolean m_IsTwiceDifferentiable
Indicates whether or not the function is twice differentiable. That is, the value returned by isTwiceDifferentiable(). Derived classes must fill this slot in their constructor.

m_IsStateless

boolean m_IsStateless
Indicates whether or not the function output depends solely of the current input (and not of the previous pattern it has processed). That is, the value returned by isStateless(). Derived classes must fill this slot in their constructor.

Class lnsc.AbstractSimpleUnit implements Serializable

Serialized Fields

m_Factor

double m_Factor
Factor property data.

m_Offset

double m_Offset
Offset property data.

Class lnsc.DataSet implements Serializable

Serialized Fields

m_Datas

java.util.Hashtable m_Datas
Hash table to store datas

Class lnsc.DataSetCollection implements Serializable

Serialized Fields

m_DataCollections

java.util.Hashtable m_DataCollections
Hash table to store datas. Elements in it are vectors

m_DataSetCount

int m_DataSetCount
Number of data set in the collection

Class lnsc.DataSetException implements Serializable

Class lnsc.InvalidDataException implements Serializable

Serialized Fields

m_DataName

java.lang.String m_DataName

Class lnsc.LinearUnit implements Serializable

Class lnsc.LogisticUnit implements Serializable

Serialized Fields

m_Alpha

double m_Alpha
Alpha property data.

m_Beta

double m_Beta
Beta property data.

m_Mu

double m_Mu
Mu property data.

Class lnsc.MissingDataException implements Serializable

Serialized Fields

m_DataName

java.lang.String m_DataName

Package lnsc.lstm

Class lnsc.lstm.ETLSTMMemoryBlock1 implements Serializable

Serialized Fields

m_Lambda

double m_Lambda

m_OppSignResetTraces

boolean m_OppSignResetTraces

Class lnsc.lstm.ETLSTMNetwork1 implements Serializable

Class lnsc.lstm.FastLSTMMemoryBlock implements Serializable

Serialized Fields

m_MemoryCellCount

int m_MemoryCellCount
Number of memory cells in the block.

m_InputGate

FunctionalUnit m_InputGate
Input gate processing function (in).

m_ForgetGate

FunctionalUnit m_ForgetGate
Forget gate processing function (fgt).

m_OutputGate

FunctionalUnit m_OutputGate
Output gate processing function (out).

m_g

FunctionalUnit m_g
First processing function (g).

m_h

FunctionalUnit m_h
Second processing function (h).

m_MemoryCellWeights

double[][] m_MemoryCellWeights
Weights from the feeding units to the memory cells [MemoryCellCount][InputCount].

m_InputGateWeights

double[] m_InputGateWeights
Weights from the feeding units to the input gate [InputCount].

m_ForgetGateWeights

double[] m_ForgetGateWeights
Weights from the feeding units to the forget gate [InputCount].

m_OutputGateWeights

double[] m_OutputGateWeights
Weights from the feeding units to the output gate [InputCount].

m_InputGatePeepholeWeights

double[] m_InputGatePeepholeWeights
Weights from the memory cell states to the input gate [MemoryCellCount].

m_ForgetGatePeepholeWeights

double[] m_ForgetGatePeepholeWeights
Weights from the memory cell states to the forget gate [MemoryCellCount].

m_OutputGatePeepholeWeights

double[] m_OutputGatePeepholeWeights
Weights from the memory cell states to the output gate [MemoryCellCount].

Class lnsc.lstm.FastLSTMNetwork implements Serializable

Serialized Fields

m_BiasToOutput

boolean m_BiasToOutput
Indicates whether bias should be connected to the output layer.

m_InputToOutput

boolean m_InputToOutput
Indicates whether input should be connected to the output layer.

m_GateToOutput

boolean m_GateToOutput
Indicates whether gates of memory block should be connected to the output layer.

m_GateToGate

boolean m_GateToGate
Indicates whether gates of memory block should be recurrently connected to memory blocks.

m_MemoryBlockCount

int m_MemoryBlockCount
Number of memory blocks.

m_MemoryBlocks

FastLSTMMemoryBlock[] m_MemoryBlocks
Memory blocks.

m_OutputLayer

FastSingleLayerNeuralNetwork m_OutputLayer
Output layer.

m_OutputWeightsLocalGradientFactor

double m_OutputWeightsLocalGradientFactor
Output layer local gradient factor.

m_Debug

boolean m_Debug
Public debug info output.

Class lnsc.lstm.OnlineSPMSELearning implements Serializable

Serialized Fields

m_Func

FunctionalUnit2 m_Func
Network to be trained.

m_Alpha

double m_Alpha
Learning rate.

m_Index

int m_Index
Input index.

Package lnsc.page

Class lnsc.page.AbstractActionRepresentation implements Serializable

Serialized Fields

m_OutputCount

int m_OutputCount
Value returned by ActionRepresentation.getOutputCount(). No default (must be specified).

Class lnsc.page.AbstractAgent implements Serializable

Serialized Fields

m_EvalMode

boolean m_EvalMode
Value returned by Agent.getEvalMode(). Default = false.

m_IsEvaluable

boolean m_IsEvaluable
Value returned by Agent.isEvaluable(). Default = true.

m_IsAdaptive

boolean m_IsAdaptive
Value returned by Agent.isAdaptive(). Default = false.

Class lnsc.page.AbstractObservableAgent implements Serializable

Serialized Fields

m_EvalMode

boolean m_EvalMode
Value returned by Agent.getEvalMode(). Default = false.

m_IsEvaluable

boolean m_IsEvaluable
Value returned by Agent.isEvaluable(). Default = true.

m_IsAdaptive

boolean m_IsAdaptive
Value returned by Agent.isAdaptive(). Default = false.

Class lnsc.page.AbstractStateRepresentation implements Serializable

Serialized Fields

m_OutputCount

int m_OutputCount
Value returned by StateRepresentation.getOutputCount(). No default (must be specified).

m_IsStateless

boolean m_IsStateless
Value returned by StateRepresentation.isStateless(). Default is true, otherwise reset() must be override.

Package lnsc.pmvf

Class lnsc.pmvf.AbstractFunctionalUnit2 implements Serializable

Serialized Fields

m_InputCount

int m_InputCount
Indicates the number of variables of the function. That is, the value returned by getInputCount(). Derived classes must fill this slot in their constructor.

m_OutputCount

int m_OutputCount
Indicates the number of values returned by the function. That is, the value returned by getOutputCount(). Derived classes must fill this slot in their constructor.

m_IsDifferentiable

boolean m_IsDifferentiable
Indicates whether or not the function is differentiable. That is, the value returned by isDifferentiable(). Derived classes must fill this slot in their constructor.

m_IsTwiceDifferentiable

boolean m_IsTwiceDifferentiable
Indicates whether or not the function is twice differentiable. That is, the value returned by isTwiceDifferentiable(). Derived classes must fill this slot in their constructor.

m_ParameterCount

int m_ParameterCount
Indicates the number of parameters for this function. That is, the value returned by getParameterCount(). Derived classes must fill this slot in their constructor.

m_IsParameterDifferentiable

boolean m_IsParameterDifferentiable
Indicates whether or not the function is differentiable with respect to its parameters. That is, the value returned by isParameterDifferentiable(). Derived classes must fill this slot in their constructor.

m_IsParameterTwiceDifferentiable

boolean m_IsParameterTwiceDifferentiable
Indicates whether or not the function is twice differentiable with respect to its parameters. That is, the value returned by isParameterTwiceDifferentiable(). Derived classes must fill this slot in their constructor.

m_IsStateless

boolean m_IsStateless
Indicates whether or not the function output depends solely of the current input (and not of the previous pattern it has processed). That is, the value returned by isStateless(). Derived classes must fill this slot in their constructor.

Class lnsc.pmvf.FastSingleLayerNeuralNetwork implements Serializable

Serialized Fields

m_HasBias

boolean m_HasBias
Indicates whether there must be an internal bias.

m_Weights

double[][] m_Weights
Weights property data.

m_OutputUnits

FunctionalUnit[] m_OutputUnits
Output units.

m_UseLeftMultiplication

boolean m_UseLeftMultiplication
Indicates whether it used weights matrix as left multiplication (y = Wx) of the input pattern or as right multiplication (y = xW) of the inputs.

m_UseRowsConcatenation

boolean m_UseRowsConcatenation
Indicates whether the matrix is tranformed by concatenation of its rows (p = [w1 | ... | wk] where wi are the rows) or its column (p = [w1T | ... | wkT] where wj are the columns) to generate the vector parameter representation.

Package stimulusdelayreward

Class stimulusdelayreward.ActorCritic_PDAETLSTM_Monkey2 implements Serializable

Serialized Fields

m_ACMModel

Agent m_ACMModel
AC model.

m_LSTMNet

ETLSTMNetwork1 m_LSTMNet
LSTM network model.

m_Trainer

OnlineSPMSELearning m_Trainer
LSTM trainer.

m_LSTMStateRep

StateRepresentation m_LSTMStateRep
LSTM state representation.

m_ACMStateRep

StateRepresentation m_ACMStateRep
AC state representation.

m_ACMExtendedStateRep

StateRepresentation m_ACMExtendedStateRep
AC state representation with previous LSTM.

m_LSTMCount

int m_LSTMCount
LSTM total outputs count.

m_LSTMlr

double m_LSTMlr
DA2

Class stimulusdelayreward.ActorCritic_PETLSTM_Monkey1 implements Serializable

Serialized Fields

m_ACMModel

Agent m_ACMModel
AC model.

m_LSTMNet

ETLSTMNetwork1 m_LSTMNet
LSTM network model.

m_Trainer

OnlineSPMSELearning m_Trainer
LSTM trainer.

m_LSTMStateRep

StateRepresentation m_LSTMStateRep
LSTM state representation.

m_ACMStateRep

StateRepresentation m_ACMStateRep
AC state representation.

m_ACMExtendedStateRep

StateRepresentation m_ACMExtendedStateRep
AC state representation with previous LSTM.

m_LSTMCount

int m_LSTMCount
LSTM total outputs count.

Class stimulusdelayreward.FlexibleSignalStateRepresentation implements Serializable

Serialized Fields

m_Bias

boolean m_Bias
Indicates whether there is a bias signal.

m_CS

boolean m_CS
Indicates whether there is a CS signal.

m_US

boolean m_US
Indicates whether there is a US (reward) signal.

Class stimulusdelayreward.OfflineStateRepresentation implements Serializable

Class stimulusdelayreward.TwoSignalStateRepresentation implements Serializable

Overview  Package  Class  Tree  Index  Help 
 PREV   NEXT FRAMES    NO FRAMES