Optimal Defense Policies for Partially Observable Spreading Processes on Bayesian Attack Graphs.
Erik MiehlingMohammad RasouliDemosthenis TeneketzisPublished in: MTD@CCS (2015)
Keyphrases
- partially observable
- markov decision problems
- markov decision processes
- attack graphs
- reward function
- decision problems
- state space
- attack graph
- partial observability
- dynamic programming
- optimal policy
- dynamical systems
- partially observable markov decision processes
- average cost
- infinite horizon
- reinforcement learning
- action models
- decision processes
- intrusion detection
- bayesian networks
- data mining
- partial observations
- knowledge base
- network configuration
- belief state
- network security
- neural network
- optimal solution
- linear programming
- search algorithm
- probabilistic model