Summary of Benchmarking Spurious Bias in Few-shot Image Classifiers, by Guangtao Zheng et al.
Benchmarking Spurious Bias in Few-Shot Image Classifiers
by Guangtao Zheng, Wenqian Ye, Aidong Zhang
First submitted to arxiv on: 4 Sep 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
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 proposes FewSTAB, a benchmark framework for assessing the robustness of few-shot image classifiers against spurious bias. Spurious bias refers to the reliance on correlations between classes and attributes that do not exist in reality. The proposed framework creates evaluation tasks with biased attributes, eliminating the need for manual dataset curation. FewSTAB uses attribute-based sample selection strategies based on a pre-trained vision-language model. This allows for automatic benchmarking of spurious bias using any existing test data. The paper demonstrates the effectiveness of FewSTAB through experiments on ten few-shot learning methods across three datasets. This framework can inspire new designs of robust few-shot classifiers, enabling them to perform well in real-world scenarios. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps make computer vision systems more reliable. It proposes a system called FewSTAB that checks how well these systems do when they’re not trained very well. Sometimes, these systems get confused because they rely on fake patterns in the data instead of actual differences between things. The new system creates fake tasks to test how well the systems can handle these kinds of mistakes. This helps developers make better, more reliable systems. |
Keywords
* Artificial intelligence * Few shot * Language model
