Machine Learning + X Seminars 2022

December 30 Recording
December 30, 2022: Transformer for Partial Differential Equations’ Operator Learning by Oded Ovadia, Tel Aviv University
December 30, 2022: Convergence analysis of the Multi-scale Deep Neural Network (MscaleDNN) by Bo Wang, Hunan Normal University
December 23 Recording
December 23, 2022: A Dimension-Augmented Physics-Informed Neural Network (DaPINN) with High Level Accuracy and Efficiency by Kamaljyoti Nath, Brown University
December 23, 2022: Optimal Approximation Rates for Deep ReLU Neural Networks on Sobolev Spaces by Jonathan Siegel, Texas A&M University
December 16 Recording
December 16, 2022: Physics-informed neural network for charged particles surrounded by conductive boundaries by Fateme Hafezi, Institute for Advanced Studies in Basic Sciences, Iran
December 16, 2022: Reliable extrapolation of deep neural operators informed by physics or sparse observations by Min Zhu, University of Pennsylvania
December 9 Recording
December 9, 2022: The Mori-Zwanzig formulation of deep learning by Xiantao Li, Pennsylvania State University
December 9, 2022: Energetically consistent model reduction for Hamiltonian and metriplectic systems by Anthony David Gruber, Sandia National Laboratories
December 2 Recording
December 2, 2022: DASA-PINNs: Differentiable adversarial self-adaptive pointwise weighting scheme for physics-informed neural networks by Guangtao Zhang, University of Macau
December 2, 2022: On the Activation Function Dependence of the Spectral Bias of Neural Networks by Adar Kahana, Brown University
November 25 Recording
November 25, 2022: Learning to Generate, Edit, and Arrange 3D Shapes by Srinath Sridhar, Brown University
November 25, 2022: TransPolymer: a Transformer-based Language Model for Polymer Property Predictions by Changwen Xu, Carnegie Mellon University
November 18 Recording
November 18, 2022: GeONet: a neural operator for learning the Wasserstein geodesic by Xiaohui Chen, University of Illinois Urbana-Champaign
November 18, 2022: Self-Validated Physics-Embedding Network (SVPEN): A general framework for inverse problems by Ruiyuan Kang, Dimitrios Kyritsis, Khalifa University, Abu Dhabi, UAE
November 11 Recording
November 11, 2022: Challenges of neural PDE surrogates by
Max Welling & Johannes Brandstetter, Microsoft Research, Amsterdam
November 11, 2022: Towards Deep Learning for Wind Farm Wake Modeling by Suraj Pawar & Ashesh Sharma, Shell Global Solutions & National Renewable Energy Laboratory
November 4 Recording
November 4, 2022: A Review of Diffusion Models by Leonard Gleyzer, Brown University
November 4, 2022: From GRAND to GRAFF: PDE and energy-based learning algorithms for graph neural networks by Zhen Zhang, Quercus Hernandez, Brown University
October 28 Recording
October 28, 2022: Towards Responsible AI via Self-Healing Inference and Tensor-Compressed Training by Zheng Zhang, University of California, Santa Barbara
October 28, 2022: Presentation of “ A Robust Scientific Machine Learning for Optimization: A Novel Robustness Theorem ” by Queroz et al. by Jerome Darbon, Brown University
October 21 Recording
October 21, 2022: Discontinuity Computing Using Physics-Informed Neural Networks by Li Liu, Institute of Applied Physics and Computational Mathematics, China
October 21, 2022: Exact conservation laws for neural network integrators of dynamical systems by Eike Mueller, University of Bath, UK
October 14 Recording
October 14, 2022: Reconstructing the pressure field around a swimming fish using a physics-informed neural network by Michael Calicchia, John Hopkins University
October 14, 2022: Is L2 Physics Informed Loss Always Suitable for Training Physics Informed Neural Network? by Shanda Li, Carnegie Mellon University
October 7 Recording
October 7, 2022: Physics-Informed Neural Networks for Shell Structures by Jan-Hendrik Bastek, Dennis M. Kochmann, ETH Zürich
October 7, 2022: Deep learning tools tutorial: neural network intelligence (NNI) by Zongren Zou, Brown University
September 30 Recording
September 30, 2022: Can a Fruit Fly Learn Word Embeddings? by Oded Ovadia, Tel Aviv University
September 30, 2022: UCNS3D- An open-source high-order compressible flow solver for unstructured meshes: philosophy and future by Panagiotis Tsoutsanis, Cranfield University
September 23 Recording
September 23, 2022: A Hybrid Iterative Numerical Transferable Solver (HINTS) for PDEs Based on Deep Operator Network and Relaxation Methods by Enrui Zhang, Adar Kahana, Brown University
September 23, 2022: A neural network solves, explains, and generates university math problems by program synthesis and few-shot learning at human level by Qian Zhang, Brown University
September 16 Recording
September 16, 2022: Deep learning-based reduced order models for parametrized PDEs by Stefania Fresca, Politecnico di Milano
September 16, 2022: I-FENN: Integrated Finite Element Neural Network. Application to continuum damage mechanics of quasi-brittle materials by Panos Pantidis, New York University
September 9 Recording
September 9, 2022: Predicting Wind-Driven Spatial Deposition through Deep Autoencoders in a Complex Terrain by Giselle Fernandez, Lawrence Livermore National Laboratory
September 9, 2022: Physics-informed neural network forward and inverse modeling of multiphase poromechanics using stress-split sequential training by Ehsan Haghighat, Carbon3D
September 1 Recording
September1, 2022: Open Access Benchmark Datasets and Metamodels for Problems in Mechanics by Emma Lejeune, Boston University
September 1, 2022: A comprehensive study of non-adaptive and residual-based adaptive sampling for physics-informed neural networks by Chenxi Wu, Brown University
August 26 Recording
August 26, 2022: Reconstructing cell phenotypic transition dynamics from single cell data by Jianhua Xing, University of Pittsburgh
August 26, 2022: Physics-informed, noise-aware, and equality Constrained NN (Forward/Inverse Problems with Multi-fidelity data) by Shams Basir, University of Pittsburgh
August 19 Recording
August 19, 2022: PINN Training using Biobjective Optimization: The Trade-off between Data Loss and Residual Loss by Kathrin Klamroth, Matthias Ehrhardt, University of Wuppertal
August 19, 2022: Data-driven discovery of Fokker-Planck equation for the Earth’s radiation belts electrons using Physics-Informed Neural Networks by Enrico Camporeale, University of Colorado
August 12 Recording
August 12, 2022: Machine learning for Fluid Mechanics by Steven L. Brunton, University of Washington
August 12, 2022: Designing and Developing Physics-Informed Machine Learning Models and Methods for Reusing Power System Components by Federica Bragone, KTH Royal Institute of Technology
August 5 Recording
August 5, 2022: From Real-time Optimal Control to Stiff Chemical Kinetics: Merging Physics Informed Neural Networks and Theory of Functional Connections for fast and accurate solutions of ODEs by Roberto Furfaro, University of Arizona
August 5, 2022: Accelerated training of Physics-informed Neural Networks (PINNs) using Meshless Discretization by Varun Shankar, University of Utah
July 29 Recording
July 29, 2022: Recurrence CFD: Fast, data-assisted simulations of flows with multiple time scales by Thomas Lichtenegger, Johannes Kepler University
July 29, 2022: Solving Quantitative Reasoning Problems with Language Models by Qian Zhang, Brown University
July 22 Recording
July 22, 2022: Understanding the origin of El Niño in El Niño-Southern Oscillation model and the forecasting of Sea surface temperature using ensemble deep learning model by Arnob Ray, Indian Statistical Institute
July 22, 2022: Constructing robust high order entropy stable discontinuous Galerkin methods by Jesse Chan, Rice University
July 22, 2022: Learning PINNs through the Van der Pol Equation by Theodore Tellides, Brown University
July 22, 2022: Paradigm shift from multiscale modeling, UQ and fractional calculus to scientific machine learning by He Li, Brown University
July 15 Recording
July 15, 2022: How a bio-inspired olfactory learning algorithm can be implemented 200x faster using Actors instead of Spiking Neural Networks by Srinivas Vamsi Parasa, Intel Corporation
July 15, 2022: Physics Informed Adversarial Training for Solving Partial Differential Equations by Simin Shekarpaz, Brown University
July 7 Recording
July 7, 2022: Deep-learning-based short-term electricity load forecasting: A real case application by Ibrahim Yazici, Istanbul Technical University
July 7, 2022: Auto-PINN: Understanding and Optimizing Physics-Informed Neural Architecture by Yicheng Wang, Texas A&M University
July 1 Recording
July 1, 2022: Generative modeling of turbulence by Hanno Gottschalk, University of Wuppertal
July 1, 2002: Learning time-dependent PDE solver using message passing graph neural networks by Zhen Zhang, Brown University
June 24 Recording
June 24, 2022: From Neural SDEs to Neural SPDEs by Cristopher Salvi, Imperial College London
June 24, 2022: Learning Stochastic Dynamics with Statistics-Informed Neural Network by Yu-Hang Tang, Lawrence Berkeley National Laboratory
June 17 Recording
June 17, 2022: Constitutive Artificial Neural Networks: a tailor-made tool for nonlinear anisotropic elasticity in biomechanics and beyond by Christian J. Cyron, Kevin Linka, Hamburg University of Technology
June 17, 2022: Interfacing finite elements with deep neural operators for fast multiscale modeling of mechanics problems by Minglang Yin, Brown University
June 10 R ecording
June 10, 2022: Generative adversarial training of recurrent neural network for probabilistic forecast of random dynamical system by Kyongmin Yeo, IBM T.J. Watson Research Center
June 10, 2022: SVD Perspectives for Augmenting DeepONet: Flexibility and Interpretability by Simone Venturi & Tiernan Casey, Sandia National Laboratories
June 3 Recording
June 3, 2022: Molecular Contrastive Learning of Representation by Amir Barati Farimani, Carnegie Mellon University
June 3, 2022: Differential equations-related machine learning: revealing hidden dynamics and learning the solution operator of PDEs by Pipi Hu, Tsinghua University & Yanqi Lake Beijing Institute of Mathematical Sciences and Applications
May 27 Recording
May 27, 2022: Learning nonlinear dynamical systems from data using scientific machine learning by Romit Maulik, Argonne National Laboratory
May 27, 2022: Deep adaptive sampling for solving high-dimensional PDEs by Xiaoliang Wan, Louisiana State University
May 20 Recording
May 20, 2022: Multilevel PINNs (MPINNs) by Elisa Riccietti, Ecole Normale Superieure, Valentin Mercier, Serge Gratton, University of Toulouse
May 20, 2022: A new application to the universal operator approximation theorem of Deep Operator Networks, to model complex physical systems controlled by external processes by Mathieu Chalvidal, Brown University
May 12 Recording
May 12, 2022: Continuous Optimization for Learning Bayesian Networks by Yue Yu, Lehigh University
May 12, 2022: Learning-Based Actuator Placement and Allocation for Resilience against Attacks by Kyriakos Vamvoudakis, Georgia Institute of Technology
May 6 Recording
May 6, 2022: (Learned) simulation as the engine of physical scene understanding by Elias Cueto, Universidad de Zaragoza
May 6, 2022: Bi-fidelity Training of Neural Networks and Neural Operators by Alireza Doostan, Shubayan De, University of Colorado Boulder
May 6, 2022: Multifidelity Deep Operator Learning by Amanda Howard, Pacific Northwest National Laboratory
May 6, 2022: Multifidelity deep neural operators for efficient learning of partial differential equations with application to fast inverse design of nanoscale heat transport by Lu Lu, University of Pennsylvania
April 29 Recording
April 29, 2022: Finite Basis Physics-Informed Neural Networks (FBPINNs): a scalable domain decomposition approach for solving differential equations by Ben Moseley, University of Oxford
April 29, 2022: Scientific Machine Learning by Dr. Jeyan Thiyagalingam, Rutherford Appleton Laboratory, Science and Technology Facilities Council (STFC)
April 22 Recording
April 22, 2022: Neuroscience-Informed Neural Networks by Sindy Löwe, University of Amsterdam
April 22, 2022: Combining Differentiable PDE Solvers and Graph Neural Networks for Fluid Flow Prediction (Paper Review) by Aniruddha Bora, Brown University
April 15 Recording
April 15, 2022: Physics-informed Attention-based Neural Network for Solving Non-linear Partial Differential Equations by Josep Ferre, Brown University
April 15, 2022:Neural Stochastic Partial Differential Equations Resolution-Invariant Learning of Continuous Spatiotemporal Dynamicsby Zongren Zou, Brown University
April 8 Recording
April 8, 2022: Physics-informed data based neural networks for two-dimensional turbulenceby Venkatesh Gopinath (Bosch Research and Technology Center) and Kannabiran Seshasayanan (Department of Physics, Indian Institute of Technology)
April 8, 2022: Graph representation learning in hyperbolic space by Sicheng Liu, Brown University
April 1 Recording
April 1, 2022: Some mathe-physical perspectives and effective theories on deep learning by Akshunna Shaurya Dogra, Imperial College London
April 1, 2022: Function regression using Spiking DeepONet by Adar Kahana, Brown University
March 25 Recording
March 25, 2022: Physics-Informed Point Net A Deep Learning Solver for Steady-State Incompressible Flows and Thermal Fields on Multiple Sets of Irregular Geometries by Ali Kashefi, Stanford Universit y
March 25, 2022: Local approximation of operators by Hrushikesh Mhaskar, Claremont Graduate University
March 18 Recording
March 18, 2022: A New Framework for Solving Dynamical Systems by Xiu Yang, Lehigh University
March 18, 2022: Respecting Causality Is All You Need for Training Physics-informed Neural Networks by Varun Kumar, Brown University
March 11 Recording
March 11, 2022: Solving Allen-Cahn and Cahn-Hilliard Equations using the Adaptive Physics Informed Neural Networks by Colby Wight, Pacific Northwest National Laboratory
March 11, 2022:Learning mappings from iced airfoils to aerodynamic coefficients using a deep operator networks by Vivek Oomen, Brown University
March 4 Recording
March 4, 2022: Physics-guided neural networks for constitutive modeling and nonlinear multiscale simulation by Oliver Weeger, Technical University of Darmstadt
March 4, 2022: Over-parameterization in manifold based surrogates and deep neural operators by Somdatta Goswami, Brown University
February 25 Recording
February 25, 2022: Phase diagram of Stochastic Gradient Descent in high-dimensional two-layer neural networks by Bruno Loureiro, École Polytechnique Fédérale de Lausanne (EPFL)
February 25, 2022: Physics-guided neural networks for constitutive modeling and nonlinear multiscale simulation by Xueyu Zhu, University of Iowa
February 18 Recording
February 18, 2022: MaxwellNet Physics-driven deep neural network training based on Maxwell’s equations by Demetri Psaltis, Ecole Polytechnique Federale de Lausanne
February 18, 2022: Improving Out-of-Distribution Robustness via Selective Augmentation (Review) by Zongren Zou, Crunch Group
February 11 Recording
February 11, 2022: Learning Operators with Coupled Attention by Jacob and Georgios, University of Pennsylvania
February 4 Recording
February 4, 2022: A Physics-Informed A Physics-Informed Neural Network to model COVID-19 Infection and Hospitalization Scenarios Neural Network to model COVID-19 Infection and Hospitalization Scenarios by Sarah Treibert and Matthias Ehrhardt, Bergische Universität Wuppertal
February 4, 2022: Neural Networks with Inputs Based on Domain of Dependence and A Converging Sequence for Solving Conservation Laws by Yingjie Liu, Georgia Institute of Technology
January 28 Recording
January 28, 2022: Transfer Learning on Multi-Fidelity Data
by Daniel Tartakovsky, Stanford University
January 28, 2022: Unsupervised multimodal data-assimilation for AI-accelerated high-throughput experimentation by Nathaniel Albert Trask, Sandia National Laboratories
Janaury 28, 2022: Tutorial on Containers using by Khemraj Shukla, Brown University (Crunch members only)
January 21 Recording
January 21, 2022: Scientific Machine Learning by Umair Waheed, King Fahd University
January 21, 2022: Presentation will be given by Qian Zhang of Brown University
Janaury 14 Recording
January 14, 2022: Physics-informed machine learning using particle methods by Alessio Alexiadis, University of Birmingham
January 14, 2022: E(n) equivariant graph neural networks & E(n) equivariant graph normalizing flows by Zhen Zhang, Brown University
Janaury 7 Recording
January 7, 2022: Weisfeiler and Lehman Go Topological Message Passing Simplicial Networks by Yu Guang Wang, Shanghai Jiao Tong University
January 7, 2022: Hierarchical Learning to Solve Partial Differential Equations Using Physics-Informed Neural Networks Networks by Yoonsang Lee, Dartmouth College