Summary of Mining In-distribution Attributes in Outliers For Out-of-distribution Detection, by Yutian Lei et al.
Mining In-distribution Attributes in Outliers for Out-of-distribution Detection
by Yutian Lei, Luping Ji, Pei Liu
First submitted to arxiv on: 16 Dec 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Machine Learning (stat.ML)
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 approach to out-of-distribution (OOD) detection in machine learning, called structured multi-view-based out-of-distribution detection learning (MVOL), is proposed. By recognizing the intrinsic correlations between in-distribution (ID) and OOD data, MVOL incorporates ID attributes into the training process, unlike previous methods that blindly suppressed these attributes. This framework utilizes both auxiliary OOD datasets and wild datasets with noisy ID data to effectively detect OOD instances. Theoretical insights on the effectiveness of MVOL are provided, demonstrating its superiority over other approaches. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Out-of-distribution detection is crucial for reliable machine learning systems in real-world scenarios. A new method called structured multi-view-based out-of-distribution detection learning (MVOL) improves OOD detection by considering intrinsic correlations between ID and OOD data. MVOL incorporates ID attributes into the training process, unlike previous methods that ignored them. This approach uses both auxiliary OOD datasets and wild datasets with noisy ID data to detect OOD instances effectively. |
Keywords
» Artificial intelligence » Machine learning
