Summary of Robust Noisy Label Learning Via Two-stream Sample Distillation, by Sihan Bai et al.
Robust Noisy Label Learning via Two-Stream Sample Distillation
by Sihan Bai, Sanping Zhou, Zheng Qin, Le Wang, Nanning Zheng
First submitted to arxiv on: 16 Apr 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- 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 The paper proposes a novel framework for noisy label learning called Two-Stream Sample Distillation (TSSD). It aims to learn robust networks by selecting high-quality samples with clean labels. The approach consists of two modules: Parallel Sample Division (PSD) and Meta Sample Purification (MSP). PSD generates reliable positive and negative samples, while MSP mines semi-hard samples from the remaining uncertain training set. The method is evaluated on four benchmark datasets and achieves state-of-the-art results. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper helps us learn better by fixing mistakes in our labels. It has a special way to pick good examples and get rid of bad ones. This makes the learning process more reliable and accurate. The approach uses two main steps: one that picks good samples and another that corrects any remaining errors. This is tested on different datasets and performs well. |
Keywords
» Artificial intelligence » Distillation
