Summary of Disentangling Tabular Data Towards Better One-class Anomaly Detection, by Jianan Ye et al.
Disentangling Tabular Data Towards Better One-Class Anomaly Detection
by Jianan Ye, Zhaorui Tan, Yijie Hu, Xi Yang, Guangliang Cheng, Kaizhu Huang
First submitted to arxiv on: 12 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 method for tabular anomaly detection under one-class classification, which involves capturing the intrinsic correlation among attributes within normal samples. The current state-of-the-art approach relies on a learnable mask strategy with a reconstruction task, but may suffer from uniform masks that reduce the effectiveness of correlation learning. To address this issue, the authors propose an innovative method that disentangles correlated subsets (CorrSets) from normal tabular data. This pioneering effort applies the concept of disentanglement to one-class anomaly detection on tabular data. The proposed method is evaluated on 20 tabular datasets and shows a substantial performance improvement of 6.1% on AUC-PR and 2.1% on AUC-ROC compared to state-of-the-art methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper helps us better detect unusual patterns in tables by figuring out what’s normal from just one type of data. This is hard because we need to find the connections between different pieces of information within normal data. The current best method tries to learn which parts of the table are most important, but it can get stuck and not work well. To solve this problem, the authors introduce a new approach that breaks down the normal data into smaller groups based on their relationships. This is the first time someone has tried this idea for detecting unusual patterns in tables. The method is tested on 20 different datasets and does significantly better than other methods. |
Keywords
» Artificial intelligence » Anomaly detection » Auc » Classification » Mask
