METANOIA:一种终身入侵检测与调查系统,用于缓解概念漂移
METANOIA: A Lifelong Intrusion Detection and Investigation System for Mitigating Concept Drift
AI总结:
针对概念漂移导致的高误报问题,METANOIA通过增量学习提出新的检测框架,通过伪边连接、可疑状态转移、路径级节点过滤和小图构建,提升检测精度,实现30%、54%和29%的性能提升。
摘要:
随着高级持续威胁(APT)复杂性的增加,溯源数据越来越多地被用于检测。基于异常的系统因其攻击知识无关性和能够对抗零日漏洞的能力而受到关注。然而,传统的检测范式,这些范式在离线、有限大小的数据上训练,往往忽视了概念漂移——流数据分布随时间的不可预测变化。这导致了高误报率。我们提出增量学习作为一种新的范式来缓解这个问题。然而,当我们将增量学习作为新的范式整合时,我们识别出四个挑战。首先,长期运行的增量系统必须对抗灾难性遗忘(C1)并避免学习恶意行为(C2)。然后,系统需要实现精确的警报(C3)和重建攻击场景(C4)。我们提出了METANOIA,这是第一个终身检测系统,旨在缓解由于概念漂移导致的高误报。它通过伪边连接来对抗灾难性遗忘,将可疑状态转移以避免学习恶意行为,通过路径级节点过滤来实现精确警报,并通过构建小图来重建攻击场景。使用最先进的基准测试,我们证明METANOIA在窗口级、图级和节点级的精度性能分别比先前的方法提高了30%、54%和29%。
英文摘要:
As Advanced Persistent Threat (APT) complexity increases, provenance data is increasingly used for detection. Anomaly-based systems are gaining attention due to their attack-knowledge-agnostic nature and ability to counter zero-day vulnerabilities. However, traditional detection paradigms, which train on offline, limited-size data, often overlook concept drift - unpredictable changes in streaming data distribution over time. This leads to high false positive rates. We propose incremental learning as a new paradigm to mitigate this issue. However, we identify FOUR CHALLENGES while integrating incremental learning as a new paradigm. First, the long-running incremental system must combat catastrophic forgetting (C1) and avoid learning malicious behaviors (C2). Then, the system needs to achieve precise alerts (C3) and reconstruct attack scenarios (C4). We present METANOIA, the first lifelong detection system that mitigates the high false positives due to concept drift. It connects pseudo edges to combat catastrophic forgetting, transfers suspicious states to avoid learning malicious behaviors, filters nodes at the path-level to achieve precise alerts, and constructs mini-graphs to reconstruct attack scenarios. Using state-of-the-art benchmarks, we demonstrate that METANOIA improves precision performance at the window-level, graph-level, and node-level by 30%, 54%, and 29%, respectively, compared to previous approaches.
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。