Papers of the day   All papers

AI Feynman: a Physics-Inspired Method for Symbolic Regression

Comments

Jul 23 2019 Sean Carroll

Great talk here at @FQXi conference by @tegmark, on how AI can mimic physicists: find a symbolic equation that best represents a dataset.
1 replies, 62 likes


Jul 23 2019 FQXi Physics

Reference: Udrescu and Tegmark "AI Feynman: a Physics-Inspired Method for Symbolic Regression": https://arxiv.org/abs/1905.11481
2 replies, 16 likes


Jun 04 2019 Daisuke Okanohara

Symbolic regression is finding a symbolic expression that matches observation (e.g., Kepler's laws). AI Feynman uses physics inspired prior (units match, low-order polynomial, symmetry) and NN regression to search most of them successfully. https://arxiv.org/abs/1905.11481
0 replies, 14 likes


Jun 01 2019 Felix Scholkmann

AI Feynman: a Physics-Inspired Method for Symbolic Regression -> https://arxiv.org/abs/1905.11481 Greath approach! #datascience #dataanalysis #artificialintelligence #AI #data #DeepLearning https://t.co/AQH7AWIMSu
0 replies, 2 likes


Content