Linking Application Description with Efficient SIMD Code Generation for Low-Precision Signed-Integer GEMM

@inproceedings{Schindler2017LinkingAD,
  title={Linking Application Description with Efficient SIMD Code Generation for Low-Precision Signed-Integer GEMM},
  author={G{\"u}nther Schindler and Manfred M{\"u}cke and Holger Fr{\"o}ning},
  booktitle={Euro-Par Workshops},
  year={2017}
}
The need to implement demanding numerical algorithms within a constrained power budget has led to a renewed interest in lowprecision number formats. Exploration of the degrees of freedom provided both by better support for low-precision number formats on computer architectures and by the respective application domain remains a most demanding task, though. In this example, we upgrade the machine learning framework Theano and the Eigen linear algebra library to support matrix multiplication of… CONTINUE READING

References

Publications referenced by this paper.
Showing 1-10 of 14 references

Deep Residual Learning for Image Recognition

2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) • 2016
View 1 Excerpt

Enerj: Approximate data types for safe and general low-power computation

A. Sampson, W. Dietl, +3 authors D. Grossman
Proceedings of the 32nd acm sigplan conference on programming language design and implementation, ser. PLDI ’11, San Jose, California, USA: ACM, 2011. • 2011

Introducing neon - development article

——
Tech. Rep., 2009. • 2009
View 1 Excerpt

Similar Papers

Loading similar papers…