Summary of The Mirrored Influence Hypothesis: Efficient Data Influence Estimation by Harnessing Forward Passes, By Myeongseob Ko et al.
The Mirrored Influence Hypothesis: Efficient Data Influence Estimation by Harnessing Forward Passes
by Myeongseob Ko, Feiyang Kang, Weiyan Shi, Ming Jin, Zhou Yu, Ruoxi Jia
First submitted to arxiv on: 14 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: 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 In this research paper, the authors tackle the pressing issue of understanding how individual data sources affect predictions made by complex machine learning models. They propose a novel approach for estimating the influence of each training point on model outputs, which is essential for ensuring the trustworthiness of these models in various applications. To overcome the computational limitations of existing methods, they develop an efficient algorithm that can be scaled up to large datasets and models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Large machine learning models are used everywhere, but we don’t really understand how each piece of data helps or hurts their predictions. The researchers behind this study want to change that by finding a better way to see which data points matter most. They’re trying to make these models more reliable and trustworthy. |
Keywords
* Artificial intelligence * Machine learning
