The activities at the SciML Lab have received the support of the National Science Foundation (NSF), through a 3-year CISE/CCF research grant, as part of an international collaboration with the Academy of Finland. Co-PIs on the project are Ming Zhong, former SciML Lab Research Associate and now an Assistant Professor at the Illinois Institute of Technology, and Simo Särkkä, Associate Professor at Aalto University, Finland. The project will benefit ongoing collaborative projects in petroleum engineering, aerospace engineering, computational biology, materials science and engineering, nuclear engineering, and astrophysics.
TAMIDS SciML Lab Seminar Series: Felipe A.C. Viana: Prognosis Digital Twins with Hybrid Physics-Informed Neural Networks
Dr. Viana is an Assistant Professor at the University of Central Florida. He is an expert in physics-informed machine learning and probabilistic methods for scientific computation. In this talk, he will describe a physics-driven and data-driven approach to building digital twins and discuss applications in engineering problems.
TAMIDS SciML Lab Seminar Series: Yen-Hsi Richard Tsai: Numerical Wave Propagation and Parallel-in-Time Computation Aided by Deep Learning
Dr. Tsai is a Professor in the Department of Mathematics and the Oden Institute for Computational Engineering and Sciences at the University of Texas at Austin. He is an expert on scientific computation and machine learning. In this talk, he will describe an approach combining deep neural networks and parareal algorithms for modeling wave propagation.
TAMIDS SciML Lab is hosting the 2nd workshop on Scientific Machine Learning on Oct 26. The workshop is open to all members of Texas A&M virtually via Zoom (TAMU authentication required). No registration is required to attend. The workshop will comprise short invited talks from Texas A&M speakers with plenty of time for technical discussion, and a round table discussion to identify challenges and opportunities for collaborative work in SciML at Texas A&M.
Updated: TAMIDS SciML Lab Workshop: TensorDiffEq for Efficient and Scalable Physics-Informed Deep Learning
TensorDiffEq is a software package designed and developed by members of the TAMU SciML community to implement collocation-based neuralPDE solvers, data assimilation solvers, as well as parameter inference and PDE discovery. This workshop aimed to promote scientific machine learning methods within the A&M research community and get more A&M researchers started in this exciting field. Meanwhile, we are interested in engaging a local community in the future development of TensorDiffEq.
ECEN Distinguished Speaker Seminar:George Karniadakis: Physics-Informed Learning for Diverse Applications in Science and Engineering
George Karniadakis is the Charles Pitts Robinson and John Palmer Barstow Professor of Applied Mathematics and Engineering at Brown University and one of the pioneers of the field of Scientific Machine Learning. He is the senior author of a number of influential papers in the area, including the highly-cited Physically-Informed Neural Network (PINN) paper. He is going to describe DeepONets for learning nonlinear operators for identifying differential equations.
TAMIDS SciML Lab Seminar Series: Jia Zhao: Solving and Learning Phase Field Models Using Modified Physics Informed Neural Networks
Dr. Zhao is an expert on computational modeling of complex multiphase fluids. He has pioneered the application of adaptive physically-informed neural networks to phase field model PDEs.
Dr. Raissi, Assistant Professor of Applied Mathematics at the University of Colorado Boulder, will give an online talk on “Hidden Physics Models” on Wednesday June 2, 2021, 3-4pm. He is the first author of several influential papers in the area, including the highly-cited Physically-Informed Neural Network (PINN) paper. He will describe his current research work in the design of physics-aware, data-efficient learning machines and novel numerical algorithms that can seamlessly blend equations and noisy multi-fidelity data and naturally quantify uncertainty in computations.
Update: TAMIDS SciML Lab Workshop: TensorDiffEq for Efficient and Scalable Physics-Informed Deep Learning
TensorDiffEq is a software package designed and developed by members of the TAMU SciML community to implement collocation-based neuralPDE solvers, data assimilation solvers, as well as parameter inference and PDE discovery. This workshop, which took place on 5/7/2021, aimed to promote scientific machine learning methods within the A&M research community and get more A&M researchers started in this exciting field. Meanwhile, we are interested in engaging a local community in the future development of TensorDiffEq.
From April 19 to April 23, 2021, the TAMIDS SciML Lab organized a one-week-long Hackathon to explore potential applications of graph learning in materiasl design. The team consisted of two faculty members from the Department of Materials Science and Engineering, one research scientist from TAMIDS, and 8 graduate students drawn from the Department of Materials Science and Engineering, Chemistry, and Electrical and Computer Engineering. Together, they were able to modify and deploy the MatErials Graph Network (MEGNet) model, one of the best libraries of graph learning in the material science field, on Grace, the new supercomputer at Texas A&M.
Dr. Sanjay Choudhry is a Senior Director at NVIDIA and leads development in AI-driven Scientific Computing & Engineering. In this seminar, Choudhry will present his team’s work on NVIDIA SimNet, a multi-physics neural solver.
TAMIDS SciML Lab Seminar Series: Chris Rackauckas: “Stiffness: Where Deep Learning Breaks and How Scientific Machine Learning Can Fix It”
Dr. Chris Rackauckas, Applied Mathematics Instructor at MIT, Director of Modeling and Simulation at Julia Computing, and the Director of Scientific Research at Pumas-AI, will present an online seminar in the TAMIDS Scientific Machine Learning (SciML) Lab Seminar Series: “Stiffness: Where Deep Learning Breaks and How Scientific Machine Learning Can Fix It” on Wednesday April 14th, 2021, 1-2pm CST. Dr. Rackauckas is a pioneer in the SciML area, whose open source DifferentialEquations.jl Julia software is widely used in academia and industry.
The TAMIDS Scientific Machine Learning (SciML) Lab is part of a new initiative to develop knowledge, resources, and community around thematic areas of Data Science / Artificial Intelligence / Machine Learning, encompassing research, education, and outreach. TAMIDS is seeking to recruit a postdoctoral research associate to join the SciML Lab multidisciplinary team, currently comprising six faculty drawn from the Colleges of Science and Engineering, researchers from TAMIDS and Texas A&M High Performance Research Computing, and associated graduate students.
TAMIDS has made an award under its Data Science Course Development Program to SciML Lab Director Dr. Ulisses Braga-Neto to develop a new for-credit class in Scientific Machine Learning. This class, with stacked sections at the undergraduate and graduate levels, will be one of the first for-credit SciML courses in the United States.
HPRC SciML Short Course (Registration Link Inside) Instructor: Jian Tao Location: Zoom session only (registration required) Prerequisites: Julia, basic understanding of partial differential equations and numerical methods. Scientific Machine Learning (SciML) is an emerging area that brings together the fields of Machine Learning and Scientific Computation. SciML introduces scientific model constraints in Machine Learning algorithms, …
TAMIDS SciML Lab Seminar Series: Paris Perdikaris: Bridging Physical Models and Observational Data with Physics-Informed Deep Learning
Dr. Paris Perdikaris, Assistant Professor in the Department of Mechanical Engineering and Applied Mechanics at the University of Pennsylvania, will present an online talk in the TAMIDS Scientific Machine Learning (SciML) Lab Seminar Series: “Bridging Physical Models and Observational Data with Physics-Informed Deep Learning” on Wednesday March 10th, 2021, 1-2pm CST.
SciML Lab Associate Director Jian Tao co-led the “12th Unmanned Team” that ranked second overall in the the 2020 AutoDrive Challenge. The team was selected to compete in GM/SAE Autodrive Challenge II.