TrojDRL: Evaluation of Backdoor Attacks on Deep Reinforcement Learning.

Panagiota Kiourti Kacper Wardega Susmit Jha Wenchao Li

Published in: DAC (2020)

Keyphrases

reinforcement learning
function approximation
evaluation model
markov decision processes
machine learning
information systems
multi agent
state space
reinforcement learning algorithms
data sets
evaluation methods
evaluation criteria
evaluation method
optimal policy
signal processing
email
cooperative