The use of partially observable Markov decision processes to optimally implement moving target defense.
Ashley McAbeeMurali TummalaJohn C. McEachenPublished in: HICSS (2021)
Keyphrases
- partially observable markov decision processes
- moving target defense
- finite state
- planning under uncertainty
- belief state
- belief space
- dynamical systems
- reinforcement learning
- markov decision processes
- continuous state
- optimal policy
- partial observability
- dynamic programming
- decision problems
- partially observable stochastic games
- stochastic domains
- partially observable markov
- state space
- sequential decision making problems
- multi agent
- planning problems
- orders of magnitude
- dec pomdps
- partially observable domains
- partially observable
- predictive state representations
- machine learning