Efficient and Reliable Deep Learning Methods and their Scientific Applications
Videos from BIRS Workshop
Russel Caflisch, Courant Institute, NYU
Monday Jun 23, 2025 09:07 - 09:32
An Adjoint Method for Optimization of the Boltzmann Equation
Cory Hauck, Oak Ridge National Laboratory
Monday Jun 23, 2025 09:33 - 10:03
Data-driven strategies for moment closures in radiation transport
Yifei Lou, University of North Carolina at Chapel Hill
Monday Jun 23, 2025 10:31 - 11:00
Graph-Based Active Learning for Nearly Blind Hyperspectral Unmixing
Zhiqiang Cai, Purdue University
Monday Jun 23, 2025 11:01 - 11:31
Neural Networks in Scientific Computing (SciML): Basics and Challenging Questions
Stanley Osher, UCLA
Monday Jun 23, 2025 12:59 - 13:29
A Novel Approach for Solving Hamilton–Jacobi Equations with Applications to Optimal Transport
Yen-Hsi Tsai, University of Texas at Austin
Monday Jun 23, 2025 13:33 - 14:28
Deep Learning Approaches for Solving Differential Equations by Classical Convergent Numerical Schemes
Wenrui Hao, Penn State University
Monday Jun 23, 2025 14:30 - 15:10
Homotopy Training Algorithms in Scientific Machine Learning
Nhat Thanh Tran, UC Irvine
Monday Jun 23, 2025 15:31 - 16:01
Efficient Local-Global Attention Approximation
Rongjie Lai, Purdue University
Monday Jun 23, 2025 16:01 - 16:32
Unsupervised Solution Operator Learning for Mean-Field Games
Yulong Lu, University of Minnesota Twin Cities
Monday Jun 23, 2025 16:37 - 17:08
Provable in-context learning of PDEs
Adam Oberman, McGill
Tuesday Jun 24, 2025 09:01 - 09:29
AI risks and current approaches to AI safety
Wei Cai, Southern Methodist University
Tuesday Jun 24, 2025 09:34 - 10:01
Martingale deep learning for very high-dimensional quasi-linear partial differential equations and stochastic optimal controls
Tao Wang, University of Victoria
Tuesday Jun 24, 2025 11:00 - 11:30
Distributed Mode Learning
Patrick Guidotti, University of California, Irvine
Tuesday Jun 24, 2025 13:00 - 13:30
Point Clouds Analysis via Kernel Interpolation and Approximate Kernel Interpolation
Haizhao Yang, University of Maryland College Park
Tuesday Jun 24, 2025 13:30 - 14:00
Modeling and Computation in the Space of Language: Symbolic and LLM-Based Approaches
Penghang Yin, SUNY Albany
Tuesday Jun 24, 2025 14:00 - 14:30
A Finite Sample Analysis for Learning Binarized Neural Network
Xiaochuan Tian, UC San Diego
Tuesday Jun 24, 2025 14:30 - 15:00
Solving Nonlinear PDEs with Sparse Radial Basis Function Networks
Qin Li, University of Wisconsin-Madison
Tuesday Jun 24, 2025 15:30 - 16:00
Optimization over probability measure space
Zhongjian Wang, Nanyang Technological University
Tuesday Jun 24, 2025 16:00 - 16:30
Wasserstein bound for generative diffusion model under Gaussian tail assumption
Shihao Zhang, UC San Diego
Tuesday Jun 24, 2025 16:30 - 17:00
Quantization and Compression of Neural Networks with Theoretical Guarantees
Yuan Gao, Purdue University
Tuesday Jun 24, 2025 19:30 - 20:00
Self-Test Loss Functions for Data-Driven Modeling of Weak-Form Operators
Yue Yu, Lehigh University
Tuesday Jun 24, 2025 20:00 - 20:30
Nonlocal Attention Operator: Towards a Foundation Model for Physical Responses
Shuhao Cao, University of Missouri-Kansas City
Tuesday Jun 24, 2025 20:30 - 21:00
Accurate Fine-Tuning of Spatiotemporal Fourier Neural Operator for Turbulent Flows
Jinchao Xu, KAUST
Wednesday Jun 25, 2025 09:00 - 09:30
Integral Representations of Sobolev Spaces via $ReLU^k$ Activation Functions and Optimal Error Estimates for Linearized Networks
Justin Sirignano, University of Oxford
Wednesday Jun 25, 2025 09:30 - 10:00
Convergence Analysis of Neural Network Methods for Solving PDEs
Molei Tao, Georgia Institute of Technology
Wednesday Jun 25, 2025 10:30 - 11:00
Where do all the scores come from? – generation accuracy of diffusion model, and multimodal sampling via denoising annealing
Shih-Hsin Wang, University of Utah
Wednesday Jun 25, 2025 11:00 - 11:30
On the Connection and Discrepancy Between Diffusion and Flow Matching
Hongkai Zhao, Duke University
Thursday Jun 26, 2025 09:00 - 09:30
Mathematical and Computational Understanding of Neural Networks: From Representation to Learning Dynamics and From Shallow to Deep
Haomin Zhou, Georgia Institute of Technology
Thursday Jun 26, 2025 09:30 - 10:00
Parameterized Wasserstein Geometric Flow
Dario Coscia, SISSA, UvA
Thursday Jun 26, 2025 10:30 - 11:00
A Variational Bayesian Method for Sequence Models Predictions and Uncertainty Quantification
Konstantinos Spiliopoulos, Boston University
Thursday Jun 26, 2025 11:00 - 11:30
Convergence Analysis of Real-time Recurrent Learning (RTRL) for a class of Recurrent Neural Networks
Kui Ren, Columbia University
Thursday Jun 26, 2025 13:00 - 13:30
A Model-Consistent Data-Driven Computational Strategy for PDE Joint Inversion Problems
Wei Zhu, Georgia Institute of Technology
Thursday Jun 26, 2025 13:30 - 14:00
Structure-preserving machine learning and data-driven structure discovery
Qi Tang, Georgia Institute of Technology
Thursday Jun 26, 2025 14:00 - 14:30
Structure-preserving machine learning for learning dynamical systems
Nicholas Boffi, Carnegie Mellon University
Thursday Jun 26, 2025 14:30 - 15:00
Stochastic interpolants: from generative modeling to generative science and engineering
Jue Yan, Iowa State University
Thursday Jun 26, 2025 15:30 - 16:00
Conservative cell-average-based neural network method for nonlinear conservation laws
Jing Qin, University of Kentucky
Thursday Jun 26, 2025 16:00 - 16:30
Form-Finding and Physical Property Predictions of Tensegrity Structures Using Deep Neural Networks
George Stepaniants, Caltech
Thursday Jun 26, 2025 16:30 - 17:00
Learning Memory and Material Dependent Constitutive Laws
