https://astroautomata.com/paper/symbolic-neural-nets/
Discovering Symbolic Models from Deep Learning with Inductive Biases
Paper Code Interactive Notebook
Paper authors: Miles Cranmer, Alvaro Sanchez-Gonzalez, Peter Battaglia, Rui Xu, Kyle Cranmer, David Spergel, Shirley Ho
“…when does a machine learning model become knowledge? ….symbolic models also boast many advantages: they’re compact, present explicit interpretations, and generalize well…. On the other hand, deep learning proves extraordinarily efficient at learning in high-dimensional spaces, but suffers from poor generalization and interpretability. So, does there exist a way to combine the strengths of both?”