Reinforcement Learning
Gambling in a rigged casino: The adversarial multi-armed bandit problem
FOCS '95 Proceedings of the 36th Annual Symposium on Foundations of Computer Science
Hi-index | 0.00 |
In this paper we present an ongoing research to develop a distributed reinforcement learning approach for mission survivability that combines two basic strategies for mission resilience: a) mission decomposition and distribution with replication of critical components, and b) differential task allocation based on estimated level of threat. Level of threat is estimated from a locally perceived attack, or the possibility of an attack, based on threat information that is shared between similar nodes.