Summary of A Skewness-based Criterion For Addressing Heteroscedastic Noise in Causal Discovery, by Yingyu Lin et al.
A Skewness-Based Criterion for Addressing Heteroscedastic Noise in Causal Discovery
by Yingyu Lin, Yuxing Huang, Wenqin Liu, Haoran Deng, Ignavier Ng, Kun Zhang, Mingming Gong, Yi-An Ma, Biwei Huang
First submitted to arxiv on: 8 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Methodology (stat.ME); 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 The paper proposes a novel approach to causal discovery by accounting for heteroscedastic noise in real-world data. It introduces heteroscedastic symmetric noise models (HSNMs) that model the effect as a function of the cause plus independent noise following a symmetric distribution. The authors develop a novel criterion, based on the skewness of the score of the data distribution, to identify HSNMs and establish the causal direction. This criterion is computationally tractable and can be extended to multivariate settings. The paper also proposes an algorithm called SkewScore that handles heteroscedastic noise without requiring exogenous noise extraction. Empirical studies validate the effectiveness of the proposed method. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper finds a way to discover causality in data that doesn’t fit certain rules. This is important because real-world data often breaks these rules. The authors create a new type of model called heteroscedastic symmetric noise models (HSNMs) that can handle this kind of noisy data. They also develop a new way to tell which direction the causal relationship goes in. This method is easy to use and can be applied to many variables at once. The paper shows that their method works well in practice. |
