Summary of Coordinated Sparse Recovery Of Label Noise, by Yukun Yang et al.
Coordinated Sparse Recovery of Label Noise
by Yukun Yang, Naihao Wang, Haixin Yang, Ruirui Li
First submitted to arxiv on: 7 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV); 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 address label noise in real-world datasets is proposed, focusing on robust classification tasks with instance-dependent label noise. Traditional methods based on sample selection often exhibit confirmation bias, which can be mitigated by using sparse over-parameterized training (SOP). However, SOP has a technical flaw that increases generalization error due to the lack of coordination between model predictions and noise recovery. To address this, Coordinated Sparse Recovery (CSR) is introduced, which uses a collaboration matrix and confidence weights to coordinate model predictions and noise recovery, reducing error leakage. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Label noise in real-world datasets can affect how well models generalize. This study looks at robust classification tasks where label noise changes from one instance to another. Current methods based on sample selection often have biases that make them less accurate. A new approach called Coordinated Sparse Recovery (CSR) is proposed, which helps reduce these biases by coordinating model predictions and noise recovery. |
Keywords
* Artificial intelligence * Classification * Generalization
