Learn about Intel® Deep Learning Boost, also known as Vector Neural Network Instructions (VNNI), a new set of AVX-512 instructions, that are designed to deliver significantly more efficient Deep Learning (Inference) acceleration. Through this technology, I will show you how you can perform low-precision (INT8) inference much faster on hardware that support the VNNI instruction set (for example, the 2nd generation Intel Xeon Scalable processors, codenamed, Cascade Lake). In the live Jupyter notebook session, you can will be able to see the benefits of this new hardware technology.
Note: This is an advanced talk. Knowledge about Deep Learning, Inferencing and basic awareness of hardware instruction sets would be desirable.