arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2409.08111cs.LGcs.AIcs.CRcs.NI
收藏

面向网络流量分析的基于图的基础模型研究

Towards a graph-based foundation model for network traffic analysis

Universitat Politécnica de Catalunya(加泰罗尼亚理工大学)

Louis Van Langendonck, Ismael Castell-Uroz, Pere Barlet-Ros

AI总结:

针对现有网络流量分析基础模型依赖token化数据包与Transformer架构的问题,提出流级别动态时空图+自监督链路预测预训练的图基础模型方案,在三类下游少样本任务中平均性能提升6.87%,展现出规模化应用潜力。

摘要:

基础模型已在多个研究领域展现出巨大应用前景。这类模型的一个潜在应用方向是计算机网络流量分析,它们能够把握网络流量动态的复杂性,且仅需少量微调即可适配任意特定任务或网络环境。此前的方法采用分词化的十六进制层级数据包数据,并依托大型语言Transformer模型的架构。我们提出了一种全新的、高效的流级别基于图的替代方案。该方法将网络流量表示为动态时空图,采用自监督的链路预测预训练任务,在该网络图框架中捕捉空间与时间动态特征。为评估方法的有效性,我们针对三项不同的下游网络任务开展了少样本学习实验,分别为入侵检测、流量分类和僵尸网络分类。从我们预训练的基础模型微调得到的模型,相较于从零开始训练的模型,平均性能提升6.87%,证明其在预训练过程中能够有效学习通用的网络流量动态特征。这一成果表明,其大规模版本有望成为可投入使用的基础模型。

英文摘要:

Foundation models have shown great promise in various fields of study. A potential application of such models is in computer network traffic analysis, where these models can grasp the complexities of network traffic dynamics and adapt to any specific task or network environment with minimal fine-tuning. Previous approaches have used tokenized hex-level packet data and the model architecture of large language transformer models. We propose a new, efficient graph-based alternative at the flow-level. Our approach represents network traffic as a dynamic spatio-temporal graph, employing a self-supervised link prediction pretraining task to capture the spatial and temporal dynamics in this network graph framework. To evaluate the effectiveness of our approach, we conduct a few-shot learning experiment for three distinct downstream network tasks: intrusion detection, traffic classification, and botnet classification. Models finetuned from our pretrained base achieve an average performance increase of 6.87\% over training from scratch, demonstrating their ability to effectively learn general network traffic dynamics during pretraining. This success suggests the potential for a large-scale version to serve as an operational foundational model.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑