Summary of Unified View Imputation and Feature Selection Learning For Incomplete Multi-view Data, by Yanyong Huang et al.
Unified View Imputation and Feature Selection Learning for Incomplete Multi-view Data
by Yanyong Huang, Zongxin Shen, Tianrui Li, Fengmao Lv
First submitted to arxiv on: 19 Jan 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 This paper proposes a novel method called UNified view Imputation and Feature selectIon lEaRning (UNIFIER) for multi-view unsupervised feature selection (MUFS). Existing MUFS methods cannot handle incomplete data, where some samples are missing in certain views. Instead, they first impute missing values and then perform feature selection. This separation misses the opportunity to leverage local structural information from feature selection to guide imputation. UNIFIER addresses these limitations by learning similarity-induced graphs from both sample and feature spaces, adapting to local structure. It dynamically recovers missing views during feature selection, using half-quadratic minimization to weight instances and alleviate outlier impact. Comprehensive experiments show UNIFIER outperforms state-of-the-art methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to choose the most important features from many views of data when some parts of that data are missing. Right now, we have to fill in the missing pieces first, then pick the best features. But this doesn’t use all the information we have. The new method, called UNIFIER, looks at both the samples and features to find patterns. It then uses these patterns to fill in the missing parts and choose the best features. This helps make sure that the results are not affected by noisy or incorrect data. The tests show that this new method works better than other methods. |
Keywords
* Artificial intelligence * Feature selection * Unsupervised
