Research Article
Optimal Decision-Making Approach for Cyber Security Defense Using Game Theory and Intelligent Learning
Algorithm 2
Defense decision-making algorithm.
| Input: ; | | Output: Defense action d | (1) | initialize , /∗ Network state and attack-defense actions are extracted by Algorithm 1 ∗/ | (2) | /∗ Getting the current network state from Network E ∗/ | (3) | repeat: | (4) | /∗ Select defense action ∗/ | (5) | Output ; /∗ Feedback defense actions to defenders ∗/ | (6) | /∗ Get the status after the action is executed ∗/ | (7) | | (8) | | (9) | for each state-action pair except do: | (10) | | (11) | | (12) | end for/∗ Update noncurrent eligibility trace and values ∗/ | (13) | /∗ Update of ∗/ | (14) | /∗ Update track of ∗/ | (15) | | (16) | Updating average strategy based on formula (6) | (17) | Selecting the learning rate of strategies based on formula (5) | (18) | | (19) | | (20) | /∗ Update defense strategy ∗/ | (21) | | (22) | end repeat |
|