# Enhancing neural operator learning with invariants to simultaneously learn various physical mechanisms

**Authors:** Siran Li, Chong Liu, Hao Ni

PMC · DOI: 10.1093/nsr/nwae198 · National Science Review · 2024-06-06

## TL;DR

This paper introduces PIANO, a new framework that improves learning of physical mechanisms by incorporating physical knowledge into neural operators.

## Contribution

The novel Physics Invariant Attention Neural Operator (PIANO) framework is introduced for integrating physical knowledge from multi-physical scenarios.

## Key findings

- PIANO is designed to decipher and integrate physical knowledge from PDEs.
- The framework is suitable for multi-physical scenarios and sampled PDE data.

## Abstract

We discuss the recent advancement in PDE learning, focusing on Physics Invariant Attention Neural Operator (PIANO). PIANO is a novel neural operator learning framework for deciphering and integrating physical knowledge from PDEs sampled from multi- physical scenarios.

## Full-text entities

- **Genes:** ALDH7A1 (aldehyde dehydrogenase 7 family member A1) [NCBI Gene 501] {aka ATQ1, EPD, EPEO4, PDE}
- **Chemicals:** PI (-)

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC11242445/full.md

## Figures

1 figure with captions in the complete paper: https://tomesphere.com/paper/PMC11242445/full.md

## References

5 references — full list in the complete paper: https://tomesphere.com/paper/PMC11242445/full.md

---
Source: https://tomesphere.com/paper/PMC11242445