arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.09697cs.LGphysics.comp-phstat.ML
收藏

用于原子模拟的十亿参数图神经网络的训练

Towards Training Billion Parameter Graph Neural Networks for Atomic Simulations

Meta FAIR;National Energy Research Scientific Computing Center (NERSC)(美国国家能源研究科学计算中心)

Anuroop Sriram, Abhishek Das, Brandon M. Wood, Siddharth Goyal, C. Lawrence Zitnick

摘要:

最近,用于原子模拟建模的图神经网络(GNN)取得了进展,有可能彻底改变催化剂的发现,这是在应对气候变化所需的能源突破方面取得进展的关键一步。然而,已被证明对这项任务最有效的GNN是内存密集型的,因为它们对图中的高阶相互作用进行建模,例如三重态或四重态原子之间的相互作用,这使得扩展这些模型具有挑战性。在本文中,我们介绍了Graph Graphelism,这是一种将输入图分布在多个GPU上的方法,使我们能够训练具有数亿或数十亿参数的非常大的GNN。我们通过将最近提出的DimeNet++和GemNet模型的参数数量增加一个数量级来实证评估我们的方法。在大规模Open Catalyst 2020(OC20)数据集上,这些图形并行化模型导致1)S2EF任务的力MAE指标相对提高了15%,2)IS2RS任务的AFbT指标相对提高了21%,建立了新的最先进的结果。

英文摘要:

Recent progress in Graph Neural Networks (GNNs) for modeling atomic simulations has the potential to revolutionize catalyst discovery, which is a key step in making progress towards the energy breakthroughs needed to combat climate change. However, the GNNs that have proven most effective for this task are memory intensive as they model higher-order interactions in the graphs such as those between triplets or quadruplets of atoms, making it challenging to scale these models. In this paper, we introduce Graph Parallelism, a method to distribute input graphs across multiple GPUs, enabling us to train very large GNNs with hundreds of millions or billions of parameters. We empirically evaluate our method by scaling up the number of parameters of the recently proposed DimeNet++ and GemNet models by over an order of magnitude. On the large-scale Open Catalyst 2020 (OC20) dataset, these graph-parallelized models lead to relative improvements of 1) 15% on the force MAE metric for the S2EF task and 2) 21% on the AFbT metric for the IS2RS task, establishing new state-of-the-art results.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑