Summary of Modelling Sampling Distributions Of Test Statistics with Autograd, by Ali Al Kadhim and Harrison B. Prosper
Modelling Sampling Distributions of Test Statistics with Autograd
by Ali Al Kadhim, Harrison B. Prosper
First submitted to arxiv on: 3 May 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG); High Energy Physics – Experiment (hep-ex); Computation (stat.CO)
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 explores an approach to modeling conditional 1-dimensional sampling distributions using simulation-based inference methods. The method involves accurate modeling of the cumulative distribution function (cdf) of a test statistic, typically done using deep neural networks whose derivative with respect to the test statistic approximates the sampling distribution. This is compared to the probability density-ratio method, also known as the likelihood-ratio trick. Various neural network models are used and their predictive uncertainty is quantified through different methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper looks at a new way to understand how things work when we use test statistics from observations that have been squished down to a single number. They’re trying to figure out if this approach is better than another method called the likelihood-ratio trick. To do this, they’re using special kinds of neural networks that can help us understand how these test statistics are distributed. The goal is to see if this new way of thinking can be used in practice. |
Keywords
» Artificial intelligence » Inference » Likelihood » Neural network » Probability
