Off-policy actor-critic deep reinforcement learning methods for alert prioritization in intrusion detection systems.
Lalitha ChavaliAbhinav KrishnanParesh SaxenaBarsha MitraAneesh Sreevallabh ChivukulaPublished in: Comput. Secur. (2024)
Keyphrases
- reinforcement learning methods
- actor critic
- intrusion detection system
- reinforcement learning algorithms
- alert correlation
- reinforcement learning
- policy gradient
- intrusion detection
- natural actor critic
- model free
- anomaly detection
- temporal difference
- markov decision processes
- state space
- dynamical systems
- function approximation
- learning algorithm
- control problems
- approximate dynamic programming
- policy iteration
- average reward
- dynamic programming
- multi agent
- action selection
- unsupervised learning
- control strategies
- cost function
- optimal control