Intel Partners With Asus On Machine Learning-Powered PCs

Today at Computex in Taipei, Asus CEO Jerry Shen jointed Intel's senior vice president and general manager of the Client Computing Group, Gregory M. Bryant, to showcase two new machine learning-powered devices. Intel also introduced introduced the AI for PC Developer Program to provide machine learning (ML) tools and training for developers.

Project Precog And Zenbook Development Platform

Intel and Asus presented Project Precog, a dual-screen device with a full set of intelligence features, as well as a new Zenbook notebook development platform that comes with an Intel Movidius Visual Processing Unit (VPU). Developers will be able to take advantage of the low-power VPU to enhance their own apps’ ML capabilities.

Last year, after having previously acquired Movidius, Intel announced the Movidius Myriad X, an embedded VPU that focused on maximizing inference performance in a low-power environment. The Myriad X comes with vision accelerators, a Neural Compute Engine, imaging accelerators, and 16 SHAVE vector processors that are paired with a CPU for heterogeneous compute. The chip can deliver up to 4 trillions of operations per second (TOPS) with a power envelope of only 1.5W.

Intel On AI PC Developer Program

ML developers will be able to train using Intel’s new AI program using tools such as the recently announced Open Visual Inference & Neural network Optimization (OpenVINO) toolkit. The toolkit allows developers to take advantage of Intel’s CPUs, GPUs, and the Movidius VPU to enhance their applications with AI capabilities.

Developers can also learn how to use the Windows ML inference engine, introduced in the latest build of Windows 10, that can also take advantage of Intel’s AVX-512 CPU instructions, its integrated GPUs through the DirectX 12 interface, as well as the Movidius VPU.