Summary of Data Imputation by Pursuing Better Classification: a Supervised Kernel-based Method, By Ruikai Yang et al.
Data Imputation by Pursuing Better Classification: A Supervised Kernel-Based Method
by Ruikai Yang, Fan He, Mingzhen He, Kaijie Wang, Xiaolin Huang
First submitted to arxiv on: 13 May 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 framework for data imputation in machine learning is proposed, which effectively leverages supervision information to complete missing data conducive to classification. The framework operates in two stages: first, it uses labels to supervise the optimization of similarity relationships among data, represented by a kernel matrix, to enhance classification accuracy. To mitigate overfitting, a perturbation variable is introduced. In the second stage, the learned kernel matrix serves as additional supervision information to guide data imputation through regression using block coordinate descent method. The proposed method outperforms state-of-the-art imputation methods on four real-world datasets, particularly when more than 60% of features are missing. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Data imputation is an important process in machine learning that helps fill in missing feature elements for incomplete data sets. A new framework is designed to better complete missing data and improve classification accuracy. The framework works by first using labels to optimize how similar data points are, and then using this information to impute missing values. This approach seems to work well when a lot of features are missing. |
Keywords
» Artificial intelligence » Classification » Machine learning » Optimization » Overfitting » Regression
