Summary of Bayesian Evidence Estimation From Posterior Samples with Normalizing Flows, by Rahul Srinivasan et al.
Bayesian evidence estimation from posterior samples with normalizing flows
by Rahul Srinivasan, Marco Crisostomi, Roberto Trotta, Enrico Barausse, Matteo Breschi
First submitted to arxiv on: 18 Apr 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Cosmology and Nongalactic Astrophysics (astro-ph.CO); Machine Learning (cs.LG); General Relativity and Quantum Cosmology (gr-qc)
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 floZ, which is based on normalizing flows and estimates Bayesian evidence (and its numerical uncertainty) from pre-existing samples drawn from the unnormalized posterior distribution. The authors validate it on distributions with known analytical evidence, up to 15 parameter space dimensions, and compare it with state-of-the-art techniques like nested sampling and k-nearest-neighbors. The method is more robust in higher dimensions, especially when dealing with sharp features in the posterior distribution. The paper demonstrates its accuracy for a simple multivariate Gaussian up to 200 dimensions with 10^5 posterior samples. The proposed floZ method has wide applicability, such as estimating evidence from variational inference, Markov Chain Monte Carlo samples, or any other method that delivers samples and their likelihood from the unnormalized posterior density. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper introduces a new way to estimate how likely it is that certain events happened. The authors take some existing data and use a special formula called floZ to figure out what’s the chance of those events happening again. They tested this method on different types of data and compared it with two other ways of doing the same thing. This new way is really good at handling complex patterns in the data, which makes it useful for many different fields like physics, biology, or economics. |
Keywords
» Artificial intelligence » Inference » Likelihood
