Using Pareto optimization in Multi-Objective Reinforcement Learning (MORL) leads to better learning results for network defense games. This is particularly useful for network security agents, who must often balance several goals when choosing what action to take in defense of a network. If the defender knows his preferred reward distribution, the advantages of Pareto optimization can be retained by using a scalarization algorithm prior to the implementation of the MORL. In this paper, we simulate a network defense scenario by creating a multi-objective zero-sum game and using Pareto optimization and MORL to determine optimal solutions and compare those solutions to different scalarization approaches. We build a Pareto Defense Strategy Selection Simulator (PDSSS) system for assisting network administrators on decision-making, specifically, on defense strategy selection, and the experiment results show that the Satisficing Trade-Off Method (STOM) scalarization approach performs better than linear scalarization or GUESS method. The results of this paper can aid network security agents attempting to find an optimal defense policy for network security games.

The prioritization of capability gaps for weapon system of systems is the basis for design and capability planning in the system of systems development process. In order to address input information uncertainties, the prioritization of capability gaps is computed in two steps using the conditional evidential network method. First, we evaluated the belief distribution of degree of required satisfaction for capabilities, and then calculated the reverse conditional belief function between capability hierarchies. We also provided verification for the feasibility and effectiveness of the proposed method through a prioritization of capability gaps calculation using an example of a spatial-navigation-and-positioning system of systems.

Improving network security is a difficult problem that requires balancing several goals, such as defense cost and need for network efficiency, to achieve proper results. Modeling the network as a game and using optimization problems to select the best move in such a game can assist network administrators in determining an ideal defense strategy. However, most approaches for determining optimal game solutions tend to focus on either single objective games or merely scalarize the multiple objectives to a single of objective. In this paper, we devise a method for modeling network attacks in a zero-sum multi-objective game without scalarizing the objectives. We use Pareto Fronts to determine the most harmful attacks and Pareto Optimization to find the best defense against those attacks. By determining the optimal solutions through those means, we allow network administrators to make the final defense decision from a much smaller set of defense options. The included experiment uses minimum distance as selection method and compares the results with a minimax algorithm for the determination of the Nash Equilibrium. The proposed algorithm should help network administrators in search of a hands-on method of improving network security.