API Reference - Models - DeepNStepQLearning (Deep N-Step Q Network)

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

Constructors

new()

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

DeepNStepQLearning.new(nStep: number, discountFactor: number): ModelObject

Parameters:

• nStep: The number of future steps considered for credit assignment. Higher values extend the planning horizon but increase variance. Set to 1 for standard Q-learning. [Default: 3]

• discountFactor: The higher the value, the more likely it focuses on long-term outcomes. The value must be set between 0 and 1. [Default: 0.95]

Returns:

• ModelObject: The generated model object.

Inherited From

• DeepReinforcementLearningBaseModel

References

• Reinforcement Learning with Neural Network

• Reinforcement Q-Learning from Scratch in Python with OpenAI Gym