From c832bb8b5e7f22ca3c22fa292c2735b25e9d5be0 Mon Sep 17 00:00:00 2001 From: Silviu Marian Udrescu Date: Sat, 25 Apr 2020 17:57:10 -0400 Subject: [PATCH] Update README.md --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 681468e..7c3ba80 100644 --- a/README.md +++ b/README.md @@ -1,12 +1,12 @@ # AI-Feynman -This code is an improved implementation of AI Feynman: a Physics-Inspired Method for Symbolic Regression: https://arxiv.org/abs/1905.11481 +This code is an improved implementation of AI Feynman: a Physics-Inspired Method for Symbolic Regression, Silviu-Marian Udrescu and Max Tegmark (2019) [[arXiv](https://arxiv.org/abs/1905.11481)][[Science Advances](https://advances.sciencemag.org/content/6/16/eaay2631/tab-pdf)]. Differently from the original paper (described in the mentioned paper), the new code doesn't output just one possible equation to describe the data, but a Pareto frontier of possible equations. Among other advantages, this approach allows the code to be more robust against noise and give good approximations to the actual equations, in case that one can't be found. In order to get started, run compile.sh to compile the fortran files used for the brute force code. -ai_feynman_example.py contains an example of running a code on some examples (found in the example_data directory). More data files on which the code can be tested on can be found in the Feynman Symbolic Regression Database: https://space.mit.edu/home/tegmark/aifeynman.html. +ai_feynman_example.py contains an example of running a code on some examples (found in the example_data directory). More data files on which the code can be tested on can be found in the [Feynman Symbolic Regression Database](https://space.mit.edu/home/tegmark/aifeynman.html). The main function of the code, called by the user, has the following parameters: