Physics-informed machine learning: case studies for weather and climate modelling

K Kashinath, M Mustafa, A Albert… - … of the Royal …, 2021 - royalsocietypublishing.org
Machine learning (ML) provides novel and powerful ways of accurately and efficiently
recognizing complex patterns, emulating nonlinear dynamics, and predicting the spatio …

[HTML][HTML] Graph neural networks: A review of methods and applications

J Zhou, G Cui, S Hu, Z Zhang, C Yang, Z Liu, L Wang… - AI open, 2020 - Elsevier
Lots of learning tasks require dealing with graph data which contains rich relation
information among elements. Modeling physics systems, learning molecular fingerprints …

How powerful are spectral graph neural networks

X Wang, M Zhang - International conference on machine …, 2022 - proceedings.mlr.press
Abstract Spectral Graph Neural Network is a kind of Graph Neural Network (GNN) based on
graph signal filters. Some models able to learn arbitrary spectral filters have emerged …

Pure transformers are powerful graph learners

J Kim, D Nguyen, S Min, S Cho… - Advances in Neural …, 2022 - proceedings.neurips.cc
We show that standard Transformers without graph-specific modifications can lead to
promising results in graph learning both in theory and practice. Given a graph, we simply …

Graph neural networks: foundation, frontiers and applications

L Wu, P Cui, J Pei, L Zhao, X Guo - … of the 28th ACM SIGKDD Conference …, 2022 - dl.acm.org
The field of graph neural networks (GNNs) has seen rapid and incredible strides over the
recent years. Graph neural networks, also known as deep learning on graphs, graph …

Gemnet: Universal directional graph neural networks for molecules

J Gasteiger, F Becker… - Advances in Neural …, 2021 - proceedings.neurips.cc
Effectively predicting molecular interactions has the potential to accelerate molecular
dynamics by multiple orders of magnitude and thus revolutionize chemical simulations …

Understanding and extending subgraph gnns by rethinking their symmetries

F Frasca, B Bevilacqua… - Advances in Neural …, 2022 - proceedings.neurips.cc
Subgraph GNNs are a recent class of expressive Graph Neural Networks (GNNs) which
model graphs as collections of subgraphs. So far, the design space of possible Subgraph …

Representational strengths and limitations of transformers

C Sanford, DJ Hsu, M Telgarsky - Advances in Neural …, 2024 - proceedings.neurips.cc
Attention layers, as commonly used in transformers, form the backbone of modern deep
learning, yet there is no mathematical description of their benefits and deficiencies as …

Benchmarking graph neural networks

VP Dwivedi, CK Joshi, AT Luu, T Laurent… - Journal of Machine …, 2023 - jmlr.org
In the last few years, graph neural networks (GNNs) have become the standard toolkit for
analyzing and learning from data on graphs. This emerging field has witnessed an extensive …

Theory for equivariant quantum neural networks

QT Nguyen, L Schatzki, P Braccia, M Ragone, PJ Coles… - PRX Quantum, 2024 - APS
Quantum neural network architectures that have little to no inductive biases are known to
face trainability and generalization issues. Inspired by a similar problem, recent …