Summary of Dimensionality-aware Outlier Detection: Theoretical and Experimental Analysis, by Alastair Anderberg et al.
Dimensionality-Aware Outlier Detection: Theoretical and Experimental Analysis
by Alastair Anderberg, James Bailey, Ricardo J. G. B. Campello, Michael E. Houle, Henrique O. Marques, Miloš Radovanović, Arthur Zimek
First submitted to arxiv on: 10 Jan 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 Our research proposes a novel nonparametric method for identifying outliers in datasets. The approach takes into account local variations in intrinsic dimensionality, which is crucial for accurate detection. By leveraging the theory of Local Intrinsic Dimensionality (LID), we develop an estimator called DAO that calculates the expected density ratio between a query point and its nearest neighbor. This dimensionality-aware method outperforms three popular benchmarks – LOF, Simplified LOF, and kNN – on over 800 synthetic and real-world datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Our new way to find unusual data points is special because it looks at how the patterns in the data change from one place to another. This helps us spot things that are different even when they’re not very far away from what’s normal. We tested our method, called DAO, on lots of different kinds of data and found that it does a better job than some other popular ways people use to find outliers. |
Keywords
* Artificial intelligence * Nearest neighbor
