Saved in:
Bibliographic Details
Main Authors: Samaddar, Ankita, Potteiger, Nicholas, Koutsoukos, Xenofon
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.02875
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916506971930624
author Samaddar, Ankita
Potteiger, Nicholas
Koutsoukos, Xenofon
author_facet Samaddar, Ankita
Potteiger, Nicholas
Koutsoukos, Xenofon
contents Autonomous agents for cyber applications take advantage of modern defense techniques by adopting intelligent agents with conventional and learning-enabled components. These intelligent agents are trained via reinforcement learning (RL) algorithms, and can learn, adapt to, reason about and deploy security rules to defend networked computer systems while maintaining critical operational workflows. However, the knowledge available during training about the state of the operational network and its environment may be limited. The agents should be trustworthy so that they can reliably detect situations they cannot handle, and hand them over to cyber experts. In this work, we develop an out-of-distribution (OOD) Monitoring algorithm that uses a Probabilistic Neural Network (PNN) to detect anomalous or OOD situations of RL-based agents with discrete states and discrete actions. To demonstrate the effectiveness of the proposed approach, we integrate the OOD monitoring algorithm with a neurosymbolic autonomous cyber agent that uses behavior trees with learning-enabled components. We evaluate the proposed approach in a simulated cyber environment under different adversarial strategies. Experimental results over a large number of episodes illustrate the overall efficiency of our proposed approach.
format Preprint
id arxiv_https___arxiv_org_abs_2412_02875
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Out-of-Distribution Detection for Neurosymbolic Autonomous Cyber Agents
Samaddar, Ankita
Potteiger, Nicholas
Koutsoukos, Xenofon
Machine Learning
Artificial Intelligence
Cryptography and Security
Autonomous agents for cyber applications take advantage of modern defense techniques by adopting intelligent agents with conventional and learning-enabled components. These intelligent agents are trained via reinforcement learning (RL) algorithms, and can learn, adapt to, reason about and deploy security rules to defend networked computer systems while maintaining critical operational workflows. However, the knowledge available during training about the state of the operational network and its environment may be limited. The agents should be trustworthy so that they can reliably detect situations they cannot handle, and hand them over to cyber experts. In this work, we develop an out-of-distribution (OOD) Monitoring algorithm that uses a Probabilistic Neural Network (PNN) to detect anomalous or OOD situations of RL-based agents with discrete states and discrete actions. To demonstrate the effectiveness of the proposed approach, we integrate the OOD monitoring algorithm with a neurosymbolic autonomous cyber agent that uses behavior trees with learning-enabled components. We evaluate the proposed approach in a simulated cyber environment under different adversarial strategies. Experimental results over a large number of episodes illustrate the overall efficiency of our proposed approach.
title Out-of-Distribution Detection for Neurosymbolic Autonomous Cyber Agents
topic Machine Learning
Artificial Intelligence
Cryptography and Security
url https://arxiv.org/abs/2412.02875