Showing 1–1 of 1 results for author: Lopez, D T

Always be Pre-Training: Representation Learning for Network Intrusion Detection with GNNs

Authors: Zhengyao Gu, Diego Troy Lopez, Lilas Alrahis, Ozgur Sinanoglu

Abstract: Graph neural network-based network intrusion detection systems have recently demonstrated state-of-the-art performance on benchmark datasets. Nevertheless, these methods suffer from a reliance on target encoding for data pre-processing, limiting widespread adoption due to the associated need for annotated labels--a cost-prohibitive requirement. In this work, we propose a solution involving in-cont… ▽ More Graph neural network-based network intrusion detection systems have recently demonstrated state-of-the-art performance on benchmark datasets. Nevertheless, these methods suffer from a reliance on target encoding for data pre-processing, limiting widespread adoption due to the associated need for annotated labels--a cost-prohibitive requirement. In this work, we propose a solution involving in-context pre-training and the utilization of dense representations for categorical features to jointly overcome the label-dependency limitation. Our approach exhibits remarkable data efficiency, achieving over 98% of the performance of the supervised state-of-the-art with less than 4% labeled data on the NF-UQ-NIDS-V2 dataset. △ Less

Submitted 29 February, 2024; originally announced February 2024.

Comments: Will appear in the 2024 International Symposium on Quality Electronic Design (ISQED'24)