inference workload

AI, and specifically deep learning, is a breakthrough technology that is driving significant advancements in many vertical markets. Whether within the computer vision or natural language processing domain, deploying this technology at scale requires overcoming many challenges and constraints. Some use cases require low latency, while others are running in power constrained environments. In this session, we will discuss the landscape of deep neural network (DNN) inference optimization techniques, understand their pros and cons, and how they can be used together to deliver optimal inference performance.