DataPredict

API Reference - Models - DeepDoubleStateActionRewardStateActionV1 (Double Deep SARSA)

DeepDoubleStateActionRewardStateActionV1 is a neural network with reinforcement learning capabilities. It can predict any positive numbers of discrete values.

It uses Hasselt et al. (2010) version, where a single neural network is selected from two neural networks with equal probability for training.

Constructors

new()

Create new model object. If any of the arguments are nil, default argument values for that argument will be used.

DeepDoubleStateActionRewardStateAction.new(discountFactor: number): ModelObject

Parameters:

• discountFactor: The higher the value, the more likely it focuses on long-term outcomes. The value must be set between 0 and 1.

Returns:

• ModelObject: The generated model object.

Functions

setParameters()

Set model’s parameters. When any of the arguments are nil, previous argument values for that argument will be used.

DeepDoubleStateActionRewardStateAction:setParameters(discountFactor: number)

Parameters:

• discountFactor: The higher the value, the more likely it focuses on long-term outcomes. The value must be set between 0 and 1.

setModelParameters1()

Sets model parameters to be used by the model.

DeepDoubleStateActionRewardStateAction:setModelParameters1(ModelParameters1: ModelParameters, doNotDeepCopy: boolean)

Parameters:

• ModelParameters1: First model parameters to be used by the model.

• doNotDeepCopy: Set whether or not to deep copy the model parameters.

setModelParameters2()

Sets model parameters to be used by the model.

DeepDoubleStateActionRewardStateAction:setModelParameters2(ModelParameters2: ModelParameters, doNotDeepCopy: boolean)

Parameters:

• ModelParameters2: Second model parameters to be used by the model.

• doNotDeepCopy: Set whether or not to deep copy the model parameters.

getModelParameters1()

Sets model parameters to be used by the model.

DeepDoubleStateActionRewardStateAction:getModelParameters1(doNotDeepCopy: boolean): ModelParameters

Parameters:

• doNotDeepCopy: Set whether or not to deep copy the model parameters.

Returns:

• ModelParameters1: First model parameters that was used by the model.

getModelParameters2(doNotDeepCopy: boolean)

Sets model parameters to be used by the model.

DeepDoubleStateActionRewardStateAction:getModelParameters2(): ModelParameters

Parameters:

• doNotDeepCopy: Set whether or not to deep copy the model parameters.

Returns:

• ModelParameters2: Second model parameters that was used by the model.

Inherited From

• ReinforcementLearningBaseModel

References

• Double Sarsa and Double Expected Sarsa with Shallow and Deep Learning

• SARSA Reinforcement Learning

• Reinforcement Learning with Neural Network