|Téma:||Neuronové sítě pro symbolickou regresi|
|Vedoucí:||Ing. Jiří Kubalík, Ph.D.|
|Vypsáno jako:||Bakalářská práce, Semestrální projekt|
|Popis:||Symbolic regression (SR) is a technique to find a model in the form of analytic equations describing given data. Typically, it has been realized using the genetic programming method, an evolutionary algorithm that evolves solutions through a stochastic process mimicking the natural evolution.
Recently, several works appeared that represent the analytic expressions as a feed-forward network that allows for efficient gradient-based training.
The goal of this work is to design and experimentally evaluate a new method that builds on this idea to use the NN for solving SR efficiently.
|Literatura:||Costa, A. et al.: Fast Neural Models for Symbolic Regression at Scale, arXiv:2007.10784, 2021
Martius, G. and Lampert, Ch. H.: Extrapolation and learning equations, arXiv:1610.02995, 2016