int action;

if (Random.Range(0f, 1f) < ExplorationFactor)
{
    action = Random.Range(0, 4); //Exploration
}
else
{
    //Exploitation
    action = QTable[currentState].ToList().IndexOf(QTable[currentState].Max());
}

//Execute Action (Step the Environment)
var envResult = AgentMover.Move(action); 


oldQValue = QTable[currentState][action]; //Old Value in Q-Table
var nextMax = QTable[envResult.state].Max(); //Best Value for next state

//Calculate new Q-Value
var newQValue = oldQValue + LearningRate * (envResult.reward + DiscountFactor * nextMax - oldQValue);
QTable[currentState][action] = newQValue;

currentState = envResult.state;