Summary of Zero-shot Outlier Detection Via Prior-data Fitted Networks: Model Selection Bygone!, by Yuchen Shen et al.
Zero-shot Outlier Detection via Prior-data Fitted Networks: Model Selection Bygone!
by Yuchen Shen, Haomin Wen, Leman Akoglu
First submitted to arxiv on: 9 Sep 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 A novel pre-trained Foundation Model called FoMo-0D is proposed for zero-shot outlier detection on tabular data, addressing the challenge of model selection in unsupervised tasks. By training on synthetic data, FoMo-0D can directly predict outlier/inlier labels without parameter fine-tuning, eliminating the need to choose an algorithm/architecture and tune its hyperparameters for a new dataset. Experimental results demonstrate that FoMo-0D significantly outperforms 24 baselines on 57 real-world datasets, with an average inference time of 7.7 ms per sample, offering at least 7x speed-up compared to previous methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary FoMo-0D is a new way to find unusual data points without needing labeled information. This helps solve the problem of choosing which algorithm to use for outlier detection. By training on fake data, FoMo-0D can quickly identify whether new data is normal or not. Tests show that FoMo-0D works well on many real-world datasets and is much faster than other methods. |
Keywords
» Artificial intelligence » Fine tuning » Inference » Outlier detection » Synthetic data » Unsupervised » Zero shot
