Beyond the Continuum: The Importance of Quantization in Deep Learning

Quantization is a valuable process in Deep Learning of mapping continuous values to a smaller set of discrete finite values. It is a powerful technique that can significantly reduce the memory footprint and computational requirements of deep learning models, making them more efficient and easier to deploy on resource-constrained devices.

In this talk, we will explore the different types of quantization techniques that can be applied to deep learning models. In addition, we will give an overview of the Neural Network Compression Framework (NNCF) and how it complements the OpenVINO™ Toolkit to achieve outstanding performance.

• The value of quantization and different types of quantization
• How to harness NNCF with the OpenVINO™ toolkit
• A Jupyter Notebook demonstrating a neural network graph before-and-after quantization with performance comparisons