Summary of Tukey G-and-h Neural Network Regression For Non-gaussian Data, by Arthur P. Guillaumin et al.
Tukey g-and-h neural network regression for non-Gaussian data
by Arthur P. Guillaumin, Natalia Efremova
First submitted to arxiv on: 12 Nov 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG)
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 addresses non-Gaussian regression with neural networks by using the Tukey g-and-h transform, a flexible parametric transform that can introduce both skewness and kurtosis to a standard normal random variable. The authors demonstrate the efficiency of their procedure in simulated examples and apply it to a real-world dataset of global crop yield for several types of crops. They also show how to carry out a goodness-of-fit analysis between the predicted distributions and test data. The method is implemented using Pytorch and made available on Github and as a Pypi package. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper uses neural networks to predict the parameters of a special kind of distribution called Tukey g-and-h distribution. This distribution can have different shapes, like being skewed or heavy-tailed. The authors show how their method works in simulations and with real data about crop yields from around the world. They also explain how they check if their predictions are good fits to the actual data. |
Keywords
* Artificial intelligence * Regression
