MultiSTOP: Solving Functional Equations with Reinforcement Learning
Abstract
We develop MultiSTOP, a Reinforcement Learning framework for solving functional equations in physics. This new methodology is also able to find actual numerical solutions instead of bounds. We extend the original BootSTOP algorithm by adding multiple constraints derived from domain-specific knowledge, even in integral form, to improve the accuracy of the solution. We investigate a particular equation in a one-dimensional Conformal Field Theory.
Cite
Text
Trenta et al. "MultiSTOP: Solving Functional Equations with Reinforcement Learning." ICLR 2024 Workshops: AI4DiffEqtnsInSci, 2024.Markdown
[Trenta et al. "MultiSTOP: Solving Functional Equations with Reinforcement Learning." ICLR 2024 Workshops: AI4DiffEqtnsInSci, 2024.](https://mlanthology.org/iclrw/2024/trenta2024iclrw-multistop/)BibTeX
@inproceedings{trenta2024iclrw-multistop,
title = {{MultiSTOP: Solving Functional Equations with Reinforcement Learning}},
author = {Trenta, Alessandro and Bacciu, Davide and Cossu, Andrea and Ferrero, Pietro},
booktitle = {ICLR 2024 Workshops: AI4DiffEqtnsInSci},
year = {2024},
url = {https://mlanthology.org/iclrw/2024/trenta2024iclrw-multistop/}
}