用神经网络同时建模随机微分方程的解与不确定性。
Neural Measures for learning distributions of Random PDEs
- 将生成模型与物理信息神经网络结合,系统量化不确定性。
- 在随机微分方程和随机偏微分方程上实现高精度预测与不确定性控制。
- 适合做科学机器学习中带不确定性的建模任务的研究者。
科学机器学习(SciML)与不确定性量化(UQ)的融合是计算科学中的前沿方向。本文通过引入概率框架,改进了物理信息神经网络(PINNs),以更有效地建模复杂系统中的不确定性。该方法将生成建模技术与PINNs结合,在保持模型预测准确性的同时,系统性地控制前向问题中的不确定性。我们通过随机微分方程和随机偏微分方程(PDEs)的应用验证了该方法的有效性。
原文摘要 · Abstract (English)
The integration of Scientific Machine Learning (SciML) techniques with uncertainty quantification (UQ) represents a rapidly evolving frontier in computational science. This work advances Physics-Informed Neural Networks (PINNs) by incorporating probabilistic frameworks to effectively model uncertainty in complex systems. Our approach enhances the representation of uncertainty in forward problems by combining generative modeling techniques with PINNs. This integration enables in a systematic fashion uncertainty control while maintaining the predictive accuracy of the model. We demonstrate the utility of this method through applications to random differential equations and random partial differential equations (PDEs).
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。