Summary of A Comprehensive Survey on Data Augmentation, by Zaitian Wang et al.
A Comprehensive Survey on Data Augmentation
by Zaitian Wang, Pengfei Wang, Kunpeng Liu, Pengyang Wang, Yanjie Fu, Chang-Tien Lu, Charu C. Aggarwal, Jian Pei, Yuanchun Zhou
First submitted to arxiv on: 15 May 2024
Categories
- Main: Machine Learning (cs.LG)
- 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 This paper focuses on data augmentation techniques that generate artificial data by manipulating existing data samples to improve AI models’ generalization capabilities. By leveraging these techniques, models can achieve better applicability in tasks involving scarce or imbalanced datasets. The study proposes a unified taxonomy of data augmentation methods across multiple modalities, including single-wise, pair-wise, and population-wise sample data augmentation methods. This approach categorizes data augmentation methods for five common data modalities through an inductive approach, providing a comprehensive understanding of how existing data samples serve the data augmentation process. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research looks at ways to make artificial intelligence (AI) models better by creating fake data that’s similar to real data. This helps AI models work well even when they don’t have enough training data. The study creates a system that groups different methods for making fake data into categories, so it’s easier to understand how these methods work together. |
Keywords
» Artificial intelligence » Data augmentation » Generalization
