nuget
The common interface of the tabular Q-learning models.
IQTable
01 Syntax
Microsoft.VisualBasic.MachineLearning.QLearning.DataModel.IQTable
02 Properties
| Name | Overloads | Summary |
|---|---|---|
| Table | 1 | The state-action value table: the key is the environment state and the value is the Q-value vector of all of the possible actions. |
| ActionRange | 1 | The number of the possible actions of each environment state. |
| ExplorationChance | 1 | The chance of taking a random action instead of the best known action, which is the epsilon value of the epsilon-greedy strategy. |
| GammaValue | 1 | The discount factor of the future reward. |
| LearningRate | 1 | The learning rate of the Q-value update. |
03 Members
Table
The state-action value table: the key is the environment state and the value is the Q-value vector of all of the possible actions.
ActionRange
The number of the possible actions of each environment state.
ExplorationChance
The chance of taking a random action instead of the best known action, which is the epsilon value of the epsilon-greedy strategy.
GammaValue
The discount factor of the future reward.
LearningRate
The learning rate of the Q-value update.