Spatiotemporally Constrained Action Space Attacks on Deep Reinforcement Learning Agents.
Xian Yeow LeeSambit GhadaiKai Liang TanChinmay HegdeSoumik SarkarPublished in: CoRR (2019)
Keyphrases
- action space
- reinforcement learning agents
- reinforcement learning
- state space
- state abstraction
- markov decision processes
- dynamic environments
- real valued
- action selection
- multi agent environments
- function approximation
- single agent
- multi agent
- reinforcement learning methods
- transfer learning
- reinforcement learning algorithms
- temporal difference
- control problems
- stochastic processes
- optimal policy
- model free
- markov decision process
- markov decision problems
- state variables
- mobile robot
- active learning
- multi agent systems