ModifiedPolicyIterationOptions<T>

Options & Configuration · AiDotNet.Models.Options

Configuration options for Modified Policy Iteration agents.

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

Constructors

Constructor Summary
ModifiedPolicyIterationOptions()

Properties

Property Summary
double Theta { get; set; } Convergence threshold: evaluation stops once the largest value change in a sweep falls below it.
int ActionSize { get; set; } Number of discrete actions, or the dimension of the action vector for a continuous action space.
int MaxEvaluationSweeps { get; set; } Maximum number of policy-evaluation sweeps between policy improvements (the partial evaluation of modified policy iteration).
int StateSize { get; set; } Number of features in one observation of the environment state.