arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.01754cs.LGphysics.chem-ph
收藏

神经网络势中的单模型不确定性量化并未持续优于模型集成

Single-model uncertainty quantification in neural network potentials does not consistently outperform model ensembles

Massachusetts Institute of Technology(麻省理工学院);AGC Inc.(旭硝子株式会社)

Aik Rui Tan, Shingo Urata, Samuel Goldman, Johannes C. B. Dietschreit, Rafael Gómez-Bombarelli

AI总结:

本研究比较了神经网络势中单模型与集成的不确定性量化方法,发现单模型在多种指标上未能持续优于集成,集成在泛化与稳健性上仍占优。

摘要:

神经网络(NNs)常常对其预测赋予高置信度,即使对于分布外远处的点也是如此,这使得不确定性量化(UQ)成为一个挑战。当它们被用于模拟材料系统中的原子间势时,这一问题会导致产生非物理结构,从而扰乱模拟,或导致不能反映真实物理的有偏统计和动力学。可微的UQ技术能够发现新的信息性数据,并驱动主动学习循环以获得稳健的势。然而,对于原子模拟,存在多种UQ技术,包括新开发的技术,且对于给定情况哪种最有效或最合适没有明确的指导方针。在这项工作中,我们考察了多种UQ方案,以通过主动学习提高NN原子间势(NNIPs)的稳健性。特别是,我们将现有的基于集成的方法与使用单一确定性NN的策略进行比较:均值-方差估计、深度证据回归和高斯混合模型。我们探索了三个数据集,范围从域内插值学习到更外推的域外泛化挑战:rMD17、氨反转和块状二氧化硅玻璃。性能通过多个将模型误差与不确定性相关联的指标进行衡量。我们的实验表明,没有任何方法在各种指标上持续优于其他方法。集成在泛化和NNIP稳健性方面仍然更好;MVE仅对域内插值有效,而GMM在域外表现更好;证据回归尽管有前景,但在任何情况下都不是更优的替代方案。更广泛地说,成本效益高的单一确定性模型目前尚不能持续匹配或优于集成在NNIPs中的不确定性量化。

英文摘要:

Neural networks (NNs) often assign high confidence to their predictions, even for points far out-of-distribution, making uncertainty quantification (UQ) a challenge. When they are employed to model interatomic potentials in materials systems, this problem leads to unphysical structures that disrupt simulations, or to biased statistics and dynamics that do not reflect the true physics. Differentiable UQ techniques can find new informative data and drive active learning loops for robust potentials. However, a variety of UQ techniques, including newly developed ones, exist for atomistic simulations and there are no clear guidelines for which are most effective or suitable for a given case. In this work, we examine multiple UQ schemes for improving the robustness of NN interatomic potentials (NNIPs) through active learning. In particular, we compare incumbent ensemble-based methods against strategies that use single, deterministic NNs: mean-variance estimation, deep evidential regression, and Gaussian mixture models. We explore three datasets ranging from in-domain interpolative learning to more extrapolative out-of-domain generalization challenges: rMD17, ammonia inversion, and bulk silica glass. Performance is measured across multiple metrics relating model error to uncertainty. Our experiments show that none of the methods consistently outperformed each other across the various metrics. Ensembling remained better at generalization and for NNIP robustness; MVE only proved effective for in-domain interpolation, while GMM was better out-of-domain; and evidential regression, despite its promise, was not the preferable alternative in any of the cases. More broadly, cost-effective, single deterministic models cannot yet consistently match or outperform ensembling for uncertainty quantification in NNIPs.

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。

↑