Summary of Data-efficient and Robust Task Selection For Meta-learning, by Donglin Zhan and James Anderson
Data-Efficient and Robust Task Selection for Meta-Learning
by Donglin Zhan, James Anderson
First submitted to arxiv on: 11 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Optimization and Control (math.OC)
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 proposes a new meta-learning method called Data-Efficient and Robust Task Selection (DERTS) that addresses the limitations of traditional uniform approaches. DERTS selects weighted subsets of tasks from task pools based on minimizing approximation error in the meta-training stage, making it efficient for rapid training and robust against noisy labels. The algorithm can be incorporated into both gradient-based and metric-based meta-learning algorithms without requiring architecture modifications. Experiments show that DERTS outperforms existing sampling strategies in limited data budget and noisy task settings. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Meta-learning is a way to learn how to learn tasks efficiently. Usually, we think all tasks are equally important, but sometimes one task might be more important than others. This paper proposes a new way to choose which tasks to learn from, called DERTS (Data-Efficient and Robust Task Selection). It’s good at choosing the right tasks quickly and ignoring noisy or wrong information. You can use this method with different types of meta-learning algorithms without changing how they work. |
Keywords
» Artificial intelligence » Meta learning
