Summary of Perturb-and-compare Approach For Detecting Out-of-distribution Samples in Constrained Access Environments, by Heeyoung Lee et al.
Perturb-and-Compare Approach for Detecting Out-of-Distribution Samples in Constrained Access Environments
by Heeyoung Lee, Hoyoon Byun, Changdae Oh, JinYeong Bak, Kyungwoo Song
First submitted to arxiv on: 19 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); 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 The proposed MixDiff framework is an out-of-distribution (OOD) detection method designed for use with machine learning models, particularly those accessed remotely through APIs. The framework detects OOD samples by applying a perturbation to both the target sample and a similar in-distribution (ID) sample, then comparing the relative difference in model outputs. This model-agnostic approach is compatible with existing output-based OOD detection methods and enhances performance on various vision and text datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The MixDiff framework helps keep machine learning models safe by detecting out-of-distribution samples that might cause unreliable outputs. It works by changing both the target sample and a similar in-distribution sample, then comparing how much the model’s output changes. This way, you can tell if the sample is unusual or not. The framework is good at detecting OOD samples that make the model very confident, which is important. |
Keywords
» Artificial intelligence » Machine learning
