Summary of On the Fly Detection Of Root Causes From Observed Data with Application to It Systems, by Lei Zan et al.
On the Fly Detection of Root Causes from Observed Data with Application to IT Systems
by Lei Zan, Charles K. Assaad, Emilie Devijver, Eric Gaussier, Ali Aït-Bachir
First submitted to arxiv on: 9 Feb 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper introduces a novel structural causal model for threshold-based IT systems and proposes an algorithm to rapidly detect root causes of anomalies. The method is proven correct when root causes are not causally related, and an extension is proposed to relax this assumption by introducing an agent that intervenes in the system. The algorithm leverages causal discovery from offline data and uses subgraph traversal for new anomalies in online data. The paper demonstrates the superior performance of its methods on various datasets, including those generated from alternative structural causal models or real IT monitoring data. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us understand how computers work when things go wrong. It creates a special model to describe systems that have limits or thresholds. Then it shows an algorithm (like a recipe) to find the cause of problems in these systems. The method is good at finding causes even if they aren’t directly connected. To make it even better, the paper suggests adding an “agent” to help relax some rules and make the system more flexible. This can be useful for fixing issues in real-life computer systems. |
