NStepQLearningOptions<T>

Options & Configuration · AiDotNet.Models.Options

Configuration options for n-step Q-learning.

Inheritance: ModelOptions → ReinforcementLearningOptions<T> → NStepQLearningOptions<T>

Constructors

Constructor Summary
NStepQLearningOptions()

Properties

Property Summary
int ActionSize { get; set; } Number of discrete actions.
int NSteps { get; set; } Number of rewards (n) accumulated before bootstrapping from the Q-value estimate.
int StateSize { get; set; } Number of features in one observation of the environment state.