AUGEM: Automatically generate high performance Dense Linear Algebra kernels on x86 CPUs

By Qian Wang, Xianyi Zhang, Yunquan Zhang, Qing Yi. 2013

233 citations recorded by the source corpus.

Subfields: Electrical and Electronic Engineering, Hardware and Architecture, Optimization & Training

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Electrical and Electronic Engineering

1. Ali Farhadi
2. Jason Mars
3. Lingjia Tang
4. Mohammad Rastegari
5. Stephan Hoyer

In Hardware and Architecture

1. Jason Mars
2. Lingjia Tang
3. Jonathan J. Helmus
4. C. Jaroniec

In Optimization & Training

1. Ali Farhadi
2. Mohammad Rastegari
3. Hessam Bagherinezhad
4. Stephan Hoyer

Citations in the corpus

• Array programming with NumPy (2020, methodology): “backends such as OpenBLAS [19, 20] or Intel MKL optimized for the CPUs at hand.”
• Neurosurgeon: Collaborative Intelligence Between the Cloud and Mobile Edge (2017, methodology): “For the mobile CPU, we use OpenBLAS [19], a NEON-vectorized matrix multiplication library and use the 4 cores available.”
• LCNN: Lookup-Based Convolutional Neural Network (2016, methodology): “Since S is computed by a dense matrix multiplication, we are still able to use OpenBlas [43] for fast matrix multiplication.”