Instructor: Dwith Chenna, AMD 
Level: Intermediate
Prerequisites: Basics of CNN

Benefits:
This course enables the attendee to:

• Understand the fundamentals of quantization.
• Have an overview of different model quantization techniques.
• Deep dive into post-training quantization techniques for CNN.
• Perform quantization analysis: What to do when quantization fails?

Course Description:
Convolutional neural networks, widely used in computer vision tasks, require substantial computation and memory resources, making it challenging to run these models on resource-constrained devices. Quantization involves modifying CNNs to use smaller data types (e.g., switching from 32-bit floating-point values to 8-bit integer values). Quantization is an effective way to reduce the computation and memory bandwidth requirements of these models, and their memory footprints, making it easier to run them on edge devices. However, quantization does degrade the accuracy of CNNs. In this talk, we survey practical techniques for CNN quantization and share best practices, tools, and recipes to enable you to get the best results from quantization, including ways to minimize accuracy loss.

Intended Audience: Engineers working on AI/ML deployment, graduate students looking for AI/ML roles in the industry, senior-level engineers.

Dwith Chenna is a seasoned research and development professional specializing in algorithm development and optimization within the realms of computer vision, deep learning, and edge AI. With a robust background in crafting state-of-the-art, performance-critical perception systems, Chenna excels in navigating the complexities of optimizing deep learning models on resource-constrained hardware accelerators for AI. In his current role at AMD, he plays a pivotal role in facilitating the adoption of AMD’s machine learning inferencing solutions, ensuring a seamless customer onboarding experience and delivering top-tier customer satisfaction. He is instrumental in enabling end-to-end AI inferencing solution productization, targeting AMD's CPU, NPU, and embedded devices. His responsibilities encompass evaluating embedded algorithms for performance and accuracy, driving key performance metrics, and developing comprehensive onboarding materials, including use-cases, tutorials, and methodology documents. By engaging with software developers and collaborating with sales, marketing, and R&D teams, Chenna ensures the successful adoption of AMD’s solutions, prioritizing customer needs and managing escalations effectively.