Algorithmic indistinguishability yields a natural test for assessing whether experts incorporate this kind of "side

Data assimilation is a vital component in modern global medium-range weather forecasting systems to obtain the best estimation of the atmospheric state by combining the short-term forecast and observations. Recently, AI-based data assimilation approaches have attracted increasing attention for their significant advantages over traditional techniques in terms of computational consumption.

This work introduces Graph Reference Distribution Learning (GRDL), an efficient and accurate graph classification method. GRDL treats each graph's latent node embeddings given by GNN layers as a

Federated learning has become a pivotal distributed learning paradigm, involving collaborative model updates across multiple nodes with private data.

Specifically, we summarize our contributions into the following points.

In recent years, the design of specialized machine learning architectures for structured data has received a surge of interest. Of particular interest are architectures for data domains with inherent symmetries, such as permutation-invariance in graphs and sets, translation-invariance in images, and other symmetries that arise from fundamental laws of physics in scientific data.

Data attribution methods aim to quantify the influence of individual training samples on the prediction of artificial intelligence (AI) models.