arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2410.01153cs.LG
收藏

Text2PDE:面向易用物理仿真的潜变量扩散模型

Text2PDE: Latent Diffusion Models for Accessible Physics Simulation

Carnegie Mellon University(卡内基梅隆大学);Naval Nuclear Laboratory(海军核实验室)

Anthony Zhou, Zijie Li, Michael Schneier, John R Buchanan, Amir Barati Farimani

AI总结:

本文提出 Text2PDE 方法,通过网格自编码器与潜变量扩散模型实现高效物理仿真,并支持以文本提示生成 PDE 解,在精度、效率及扩展性上达到与现有神经求解器相当的水平。

摘要:

深度学习的最新进展激发了许多针对偏微分方程(PDE)问题的数据驱动求解方法研究。这些神经 PDE 求解器通常比数值方法快得多;然而,每种方法都有其独特的局限性,并且通常需要在训练成本、数值精度以及适用于不同问题设置的易用性之间进行权衡。为了解决这些局限性,我们提出了若干将潜变量扩散模型应用于物理仿真的方法。首先,我们引入了一种网格自编码器来压缩任意离散化的 PDE 数据,从而能够在各种物理问题上进行高效的扩散训练。此外,我们研究了完整时空解的生成,以缓解自回归误差累积。最后,我们研究了以初始物理量为条件,以及仅以文本提示为条件来实现 text2PDE 生成。我们表明,语言可以作为一种紧凑、可解释且准确的模态来生成物理仿真,为更易用、更易获取的 PDE 求解器铺平道路。通过在均匀网格和结构化网格上的实验,我们表明所提出的方法在精度和效率上均能与当前的神经 PDE 求解器相竞争,并在扩展到约 30 亿参数时表现出良好的扩展行为。通过引入一种可扩展、准确且易用的物理仿真器,我们希望使神经 PDE 求解器更接近实际应用。

英文摘要:

Recent advances in deep learning have inspired numerous works on data-driven solutions to partial differential equation (PDE) problems. These neural PDE solvers can often be much faster than their numerical counterparts; however, each presents its unique limitations and generally balances training cost, numerical accuracy, and ease of applicability to different problem setups. To address these limitations, we introduce several methods to apply latent diffusion models to physics simulation. Firstly, we introduce a mesh autoencoder to compress arbitrarily discretized PDE data, allowing for efficient diffusion training across various physics. Furthermore, we investigate full spatio-temporal solution generation to mitigate autoregressive error accumulation. Lastly, we investigate conditioning on initial physical quantities, as well as conditioning solely on a text prompt to introduce text2PDE generation. We show that language can be a compact, interpretable, and accurate modality for generating physics simulations, paving the way for more usable and accessible PDE solvers. Through experiments on both uniform and structured grids, we show that the proposed approach is competitive with current neural PDE solvers in both accuracy and efficiency, with promising scaling behavior up to $\sim$3 billion parameters. By introducing a scalable, accurate, and usable physics simulator, we hope to bring neural PDE solvers closer to practical use.

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。

↑