arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

University of Pennsylvania(宾夕法尼亚大学)

2026-05-18 至 2026-05-18 共收录 4
2505.18511 2026-05-18 cs.LG math.AP physics.comp-ph

SPDEBench: An Extensive Benchmark for Learning Stochastic PDEs

SPDEBench:学习随机偏微分方程的广泛基准

Yuantu Zhu, Zheyan Li, Dai Shi, Luke Thompson, Oliver Nash, Jose Miguel Lara Rangel, Siran Li, Bingguang Chen, Rongchan Zhu, Qi Meng, Hao Ni

机构 * Shanghai Jiao Tong University(上海交通大学) University of Pennsylvania(宾夕法尼亚大学) University of Cambridge(剑桥大学) University of Sydney(悉尼大学) Imperial College London(伦敦帝国理工学院) University College London(伦敦大学学院) Fujian Normal University(福建师范大学) Beijing Institute of Technology(北京理工大学) Chinese Academy of Sciences(中国科学院)

AI总结 本文提出SPDEBench,首个统一的ML学习随机偏微分方程基准,提供物理和数学重要的1-3维领域数据集,涵盖正则和奇异SPDE,并包含7种评估指标,验证模型精度、鲁棒性和分布外泛化能力。

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.13350 2026-05-18 cs.RO

Approximating Global Contact-Implicit MPC via Sampling and Local Complementarity

通过采样和局部互补性近似全局接触-隐式MPC

Sharanya Venkatesh, Bibit Bianchini, Alp Aydinoglu, William Yang, Michael Posa

机构 * GRASP Laboratory at the University of Pennsylvania(宾夕法尼亚大学GRASP实验室) Boston Dynamics(波士顿动力) Amazon Robotics(亚马逊机器人技术)

AI总结 本文提出一种结合局部互补性控制与全局采样方法的控制器,用于实时灵活操作。通过在每个控制循环中先进行无接触阶段再进行接触密集阶段,实现对非凸物体的精确非抓取操作。

Comments S.V. and B.B. contributed equally to this work. Accepted to RA-L 2025; presented at ICRA 2026. Project page: https://approximating-global-ci-mpc.github.io

Journal ref IEEE Robotics and Automation Letters, volume 10, number 11, pages 12117-12124, September 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.01990 2026-05-18 cs.LG cs.CE

Deep Learning Alternatives of the Kolmogorov Superposition Theorem

Kolmogorov超位置定理的深度学习替代方案

Leonardo Ferreira Guilhoto, Paris Perdikaris

机构 * Graduate Group in Applied Mathematics and Computational Science(应用数学与计算科学联合研究生组) University of Pennsylvania(宾夕法尼亚大学) Department of Mechanical Engineering & Applied Mechanics(机械工程与应用力学系)

AI总结 本文探讨了Kolmogorov超位置定理的替代形式,提出ActNet模型克服原有形式的缺陷,在PINNs框架中表现优异,适用于科学计算和PDE模拟。

Journal ref Guilhoto, Leonardo Ferreira, and Paris Perdikaris. "Deep Learning Alternatives Of The Kolmogorov Superposition Theorem." The Thirteenth International Conference on Learning Representations (ICLR 2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2404.03099 2026-05-18 cs.LG cs.AI cs.CE cs.IT math.IT stat.ML

Composite Bayesian Optimization In Function Spaces Using NEON -- Neural Epistemic Operator Networks

在函数空间中使用NEON进行复合贝叶斯优化

Leonardo Ferreira Guilhoto, Paris Perdikaris

机构 * Graduate Group in Applied Mathematics and Computational Science(应用数学与计算科学联合研究生组) University of Pennsylvania(宾夕法尼亚大学) Department of Mechanical Engineering and Applied Mechanics(机械工程与应用力学系)

AI总结 本文提出NEON网络,通过单一运算符网络 backbone 实现预测不确定性,用于解决复合贝叶斯优化问题,展示其在toy和现实场景中优于其他方法。

Journal ref Guilhoto, Leonardo Ferreira, and Paris Perdikaris. "Composite Bayesian optimization in function spaces using NEON - Neural Epistemic Operator Networks." Scientific Reports 14.1 (2024): 29199

详情

展开后加载摘要…

URL PDF HTML 收藏