Sort by
Refine Your Search
-
Listed
-
Country
-
Employer
- Argonne
- University of North Carolina at Chapel Hill
- European Space Agency
- Nature Careers
- Duke University
- Imperial College London
- Oak Ridge National Laboratory
- Stony Brook University
- Technical University of Denmark
- University of Luxembourg
- Yale University
- ; University of Copenhagen
- Brookhaven Lab
- Durham University
- Embry-Riddle Aeronautical University
- Empa
- European Magnetism Association EMA
- Harvard University
- Heriot Watt University
- MOHAMMED VI POLYTECHNIC UNIVERSITY
- NEW YORK UNIVERSITY ABU DHABI
- New York University
- Northeastern University
- Shanghai Jiao Tong University
- Stanford University
- Technical University of Munich
- The Ohio State University
- The University of Arizona
- UNIVERSITY OF HELSINKI
- University of Houston Central Campus
- University of Lund
- University of Minnesota
- University of Minnesota Twin Cities
- University of North Texas at Dallas
- University of Oxford
- University of South Carolina
- University of Texas at Arlington
- VIB
- 28 more »
- « less
-
Field
-
models (e.g., CNNs, diffusion models, etc) Proficiency in Python Experience with HPC (CPU or GPU, with GPUs preferred) Related Skills and Other Requirements Ability to collaborate on the application of AI
-
optimizing PIC algorithms for modern heterogeneous architectures, including CPUs, GPUs, and other accelerators, the project seeks to achieve unprecedented efficiency and resolution in plasma simulations
-
optimizing PIC algorithms for modern heterogeneous architectures, including CPUs, GPUs, and other accelerators, the project seeks to achieve unprecedented efficiency and resolution in plasma simulations
-
techniques. Preferred Qualifications: Knowledge of HPC matrix, tensor and graph algorithms. Knowledge of GPU CUDA and HIP programming Knowledge on distributed algorithms using MPI and other frameworks such as
-
needs, such as assisting the team with evaluating evolutionary algorithms for exploring creative new hand designs, or reinforcement learning for policy optimisation, all within a huge GPU-based simulation
-
Mathematica and Python with an interest in GPU programming. These required and desired skills should be demonstrated by presenting an existing body of code and/or peer-reviewed publications. Additional
-
libraries for modern architectures (e.g., GPUs). Exploration of linear algebra methods in computational physics applications and machine learning. Integrate and benchmark the GINGKO library, a sparse solver
-
for modern heterogeneous architectures, including CPUs, GPUs, and other accelerators, the project seeks to achieve unprecedented efficiency and resolution in plasma simulations. This advancement will enable
-
, Probabilistic Inference, Algebraic Topology and Wavelet analysis theory. Familiar with Matlab/Python/C++ programming. Experience with Pytorch and multi-GPU model deployment. Experience in analyzing complex
-
. First, efficient and scalable training procedure are still needed, irrespective of whether the training is done off-line on a traditional GPU-based architecture, on neuromorphic hardware. Second