Network Security Defense Decision-Making Method Based on Stochastic Game and Deep Reinforcement Learning

Table 4

The meanings of the math notations.

Notation	Meaning
	The state of the network system
	Attacker defender’s action sets
	The action of attacker defender
	Type of attacker
	Prior probability of attacker type
	The strategy of attacker defender
	Nash equilibrium strategy
	Probability of attacker defender choosing action
	The “state-action” revenue function of attacker defender
	The “state-strategy” revenue function of attacker defender
	Current reward
	Discount factor
	Learning rate