Deep reinforcement learning for FlipIt security game

Files
2002.12909.pdf(1.03 MB)
Accepted manuscript
Date
2021-12-02
Authors
Greige, Laura
Chin, Sang
Version
Accepted manuscript
OA Version
Citation
Greige, L., Chin, P. (2022). Deep Reinforcement Learning for FlipIt Security Game. In: Benito, R.M., Cherifi, C., Cherifi, H., Moro, E., Rocha, L.M., Sales-Pardo, M. (eds) Complex Networks & Their Applications X. COMPLEX NETWORKS 2021. Studies in Computational Intelligence, vol 1015. Springer, Cham. https://doi.org/10.1007/978-3-030-93409-5_68
Abstract
Reinforcement learning has shown much success in games such as chess, backgammon and Go [21,24,22]. However, in most of these games, agents have full knowledge of the environment at all times. In this paper, we describe a deep learning model in which agents successfully adapt to different classes of opponents and learn the optimal counter-strategy using reinforcement learning in a game under partial observability. We apply our model to FlipIt [25], a two-player security game in which both players, the attacker and the defender, compete for ownership of a shared resource and only receive information on the current state of the game upon making a move. Our model is a deep neural network combined with Q-learning and is trained to maximize the defender’s time of ownership of the resource. Despite the noisy information, our model successfully learns a cost-effective counter-strategy outperforming its opponent’s strategies and shows the advantages of the use of deep reinforcement learning in game theoretic scenarios. We also extend FlipIt to a larger action-spaced game with the introduction of a new lower-cost move and generalize the model to n-player FlipIt.
Description
License