No abstract

Graph motifs are significant subgraph patterns occurring frequently in graphs, and they play important roles in representing the whole graph characteristics. For example, in chemical domain, functional groups are motifs that can determine molecule properties. Mining and utilizing motifs, however, is a non-trivial task for large graph datasets. Traditional motif discovery approaches rely on exact counting or statistical estimation, which are hard to scale for large datasets with continuous and high-dimension features. In light of the significance and challenges of motif mining, we propose MICRO-Graph: a framework for MotIf-driven Contrastive leaRning Of Graph representations to: 1) pre-train Graph Neural Networks (GNNs) in a self-supervised manner to automatically extract motifs from large graph datasets; 2) leverage learned motifs to guide the contrastive learning of graph representations, which further benefit various downstream tasks. Specifically, given a graph dataset, a motif learner cluster similar and significant subgraphs into corresponding motif slots. Based on the learned motifs, a motif-guided subgraph segmenter is trained to generate more informative subgraphs, which are used to conduct graph-to-subgraph contrastive learning of GNNs. By pretraining on ogbg-molhiv molecule dataset with our proposed MICRO-Graph, the pre-trained GNN model can enhance various chemical property prediction downstream tasks with scarce label by 2.0%, which is significantly higher than other state-of-the-art self-supervised learning baselines.Recently, Graph Neural Networks (GNNs) have shown great expressive power for learning graph representations without explicit feature engineering (

We propose a MOTIF-driven contrastive framework to pretrain a graph neural network in a self-supervised manner so that it can automatically mine motifs from large graph datasets. Our framework achieves state-of-the-art results on various graph-level downstream tasks with few labels, like molecular property prediction.

Evaluating bias, fairness, and social impact in monolingual language models is a difficult task. This challenge is further compounded when language modeling occurs in a multilingual context. Considering the implication of evaluation biases for large multilingual language models, we situate the discussion of bias evaluation within a wider context of social scientific research with computational work. We highlight three dimensions of developing multilingual bias evaluation frameworks: (1) increasing transparency through documentation, (2) expanding targets of bias beyond gender, and (3) addressing cultural differences that exist between languages. We further discuss the power dynamics and consequences of training large language models and recommend that researchers remain cognizant of the ramifications of developing such technologies.

Stellar age cannot be directly measured, yet age determinations are fundamental to understanding the evolution of stars, planets, and galaxies. The work presented here builds upon the idea of a stellar-activity age. We utilized far-ultraviolet (FUV) photometry acquired by the Galaxy Evolution Explorer (GALEX) space telescope as an indicator of chromospheric activity to infer ages of late-F, G, and K type dwarf stars. We derived a purely empirical correlation between FUV magnitudes and stellar age in conjunction with (B − V) color. Our attention is restricted to Sun-like stars with color range and absolute magnitude range 4.3 ≤ M V ≤ 5.3. The correlation is defined in terms of a FUV-excess parameter . We related stellar age, τ, to Q through the relation , where a and b are fit parameters and functions of (B − V). This correlation is functional up to 6 Gyr for FGK dwarfs. With such a correlation, one only needs Johnson (B − V) and FUV measurements to estimate the stellar age for Population i dwarf stars of solar-like temperature and metallicity. Such a calibration has utility in population studies of FGK dwarfs for further understanding of the chemical evolution of the Milky Way. As an illustration of one such application, we have constructed activity and FUV–age distributions for a sample of thin and thick disk stars, as distinguished by their chemical abundances. Considerable overlap is found between the activity distribution and age range of the two populations. We discuss the possibility that some high-[α/Fe] thick disk stars were formed as a result of the accretion of dwarf galaxies as recently as 4 Gyr ago.