Mixed precision algorithms in numerical linear algebra
Today's floating-point arithmetic landscape is broader than ever. While scientific computing
has traditionally used single precision and double precision floating-point arithmetics, half …
has traditionally used single precision and double precision floating-point arithmetics, half …
Toward accelerated stencil computation by adapting tensor core unit on gpu
The Tensor Core Unit (TCU) has been increasingly adopted on modern high performance
processors, specialized in boosting the performance of general matrix multiplication …
processors, specialized in boosting the performance of general matrix multiplication …
Predicting Performance and Accuracy of Mixed-Precision Programs for Precision Tuning
Y Wang, C Rubio-González - Proceedings of the 46th IEEE/ACM …, 2024 - dl.acm.org
A mixed-precision program is a floating-point program that utilizes different precisions for
different operations, providing the opportunity of balancing the trade-off between accuracy …
different operations, providing the opportunity of balancing the trade-off between accuracy …
Simulation of Quantum Computers: Review and Acceleration Opportunities
Quantum computing has the potential to revolutionize multiple fields by solving complex
problems that can not be solved in reasonable time with current classical computers …
problems that can not be solved in reasonable time with current classical computers …
Neural network precision tuning using stochastic arithmetic
Q Ferro, S Graillat, T Hilaire, F Jézéquel… - … Workshop on Numerical …, 2022 - Springer
Neural networks can be costly in terms of memory and execution time. Reducing their cost
has become an objective, especially when integrated in an embedded system with limited …
has become an objective, especially when integrated in an embedded system with limited …
BinFPE: accurate floating-point exception detection for GPU applications
When modern heterogeneous HPC systems perform numerical computations, floating-point
exceptional quantities such as NaN and infinity in the GPU context, remain insufficiently …
exceptional quantities such as NaN and infinity in the GPU context, remain insufficiently …
FPChecker: Floating-point exception detection tool and benchmark for parallel and distributed hpc
Floating-point arithmetic is fundamental to many areas including high-performance
computing and machine learning. In order to ensure the numerical integrity of the overall …
computing and machine learning. In order to ensure the numerical integrity of the overall …
Towards a SYCL API for Approximate Computing
L Carpentieri, B Cosenza - … of the 2023 International Workshop on …, 2023 - dl.acm.org
Approximate computing is a well-known method [7] to achieve higher performance or lower
energy consumption while accepting a loss of output accuracy. Many applications such as …
energy consumption while accepting a loss of output accuracy. Many applications such as …
Convergence-aware operator-wise mixed-precision training
With the support of more precision formats in emerging hardware architectures, mixed-
precision has become a popular approach to accelerate deep learning (DL) training …
precision has become a popular approach to accelerate deep learning (DL) training …
MixPert: Optimizing Mixed-Precision Floating-Point Emulation on GPU Integer Tensor Cores
Z Lin, A Sun, X Zhang, Y Lu - Proceedings of the 25th ACM SIGPLAN …, 2024 - dl.acm.org
Featuring mixed-precision tensor operations, accelerators significantly enhance
performance for many error-tolerant computing tasks, but their applicability is limited in …
performance for many error-tolerant computing tasks, but their applicability is limited in …