Policy Resilience to Environment Poisoning Attacks on Reinforcement Learning.
Hang XuXinghua QuZinovi RabinovichPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- optimal policy
- real time
- markov decision process
- exploration strategy
- policy search
- agent receives
- agent learns
- multi agent environments
- markov decision problems
- reward function
- action selection
- markov decision processes
- dynamic environments
- decision problems
- control policy
- continuous state
- autonomous agents
- state space
- mobile robot
- partially observable environments
- learning process