We present MegaBlocks, a system for efficient Mixture-of-Experts (MoE) training on GPUs. Our system ismotivated by the limitations of current frameworks, which restrict the dynamic …
Tensor algebra is a powerful tool with applications in machine learning, data analytics, engineering and the physical sciences. Tensors are often sparse and compound operations …
Z Wen, J Shi, Q Li, B He, J Chen - Journal of Machine Learning Research, 2018 - jmlr.org
Support Vector Machines (SVMs) are classic supervised learning models for classification, regression and distribution estimation. A survey conducted by Kaggle in 2017 shows that …
W Liu, B Vinter - Proceedings of the 29th ACM on International …, 2015 - dl.acm.org
Sparse matrix-vector multiplication (SpMV) is a fundamental building block for numerous applications. In this paper, we propose CSR5 (Compressed Sparse Row 5), a new storage …
Phenomenally successful in practical inference problems, convolutional neural networks (CNN) are widely deployed in mobile devices, data centers, and even supercomputers. The …
M Zhu, T Zhang, Z Gu, Y Xie - Proceedings of the 52nd Annual IEEE …, 2019 - dl.acm.org
Deep neural networks have become the compelling solution for the applications such as image classification, object detection, speech recognition, and machine translation …
A Buluç, JR Gilbert - The International Journal of High …, 2011 - journals.sagepub.com
This paper presents a scalable high-performance software library to be used for graph analysis and data mining. Large combinatorial graphs appear in many applications of high …
The symmetric sparse matrix-vector multiplication (SymmSpMV) is an important building block for many numerical linear algebra kernel operations or graph traversal applications …
N Srivastava, H Jin, S Smith, H Rong… - … Symposium on High …, 2020 - ieeexplore.ieee.org
Tensor factorizations are powerful tools in many machine learning and data analytics applications. Tensors are often sparse, which makes sparse tensor factorizations memory …