Identifying the mathematical relationships that best describe a dataset
remains a very challenging problem in machine learning, and is known as
symbolic regression (SR). In contrast to neural networks which are o
Symbolic regression is the task of learning a mathematical expression for data and while historically it has been tackled by heuristics, no proof has yet been given that it is NP-hard; however, this paper provides evidence that this may be the case, indicating that an exact polynomial-time algorithm to compute SR models is unlikely.