David-Wyatt.jpg

David Wyatt

CTO, PixelDisplay

Program Chair, AI Hardware Series
Chair, AI Hardware at the Edge Track

Welcome everyone to the 2020 ValleyML AI Hardware Conference series, focused on Edge & Low-Power Hardware Solutions for AI


Introduction

No-one doubts the ability of focused hardware to make ever more amazing improvements in performance. But with each new AI chip, one of the key themes emerging has been power-efficiency. Nowhere is this more important than at the Edge, where battery-powered AI applications have the greatest potential for proliferation (e.g. growth is projected at 21~37% CAGR through 2027). But as we also see from analysis of contemporary GPU solutions, the current state-of-the-art is un-scaleable, even in the Data Center (e.g. MIT analysis of training Transformer-Model on NVIDIA GPU produced more CO2 than 5 cars in their lifetime including all the petrol consumed). For a sustainable future for AI, technology solutions for efficiency and low-power are critically needed - everywhere.

The Edge/Low-Power conference series includes Keynotes from Intel : leading the field in the development of power-efficient AI processors. And also Sandia Labs : leaders in the real-world applications of AI, sharing their views of the way forwards (including SNN, ANN, Neuromorphic, Analog, In-memory-compute). Also a first time presentation on a key missing element : Compiler designed for radical new compute-in-memory hardware AI architectures.

We’re also throwing down the gauntlet, and challenging rosy notions of some alternative architectures, with an open discussion Panel, comprising: AI Analyst and industry experts, to dig deeper and ask the tough question: “So where are all the sub-1W AI Chips ?”. The series will also include a panel featuring Silicon Valley VC’s with an open forum for Hardware Startups to come pitch their ideas.

Join us to find out more, ask the tough questions, show creative solutions, and help inspire innovation in better efficiency in the future Edge and Low-Power AI Hardware.


Chairperson Bio

David Wyatt is the CTO of PixelDisplay, a company he co-founded in 2015. David is the inventor of PixelDisplay’s VividColor™, NoBlue™, NanoBright™, and SmartPixels™ technologies that give system developers better visual solutions, and leading-edge intelligent MicroLED. Prior to founding PixelDisplay, David was previously Distinguished Engineer at NVIDIA for 9yrs, leading technology developments on GPU’s, Tegra, including inventing NVIDIA’s Hybrid Graphics, G-SYNC technologies. Before NVIDIA, David was a Chief Engineer, and Platform Architect at Intel for 8yrs; and previous lead engineering roles in early fabless graphics pioneers including Weitek and Chips&Technologies. David originates from rural Australia, the country where he studied Electrical Engineering and Computer Science at the University of Queensland, before following his engineer passion for both hardware and software into entrepreneurism, to found his first Startup in Taiwan, and coming to the US through an acquisition.

After 20 years’ experience in Silicon Valley, and more than 116 issued US patents, David’s many pioneering innovations have covered a wide range of disruptive solutions, becoming an integral part of industry standards, and the key features on shipping products including those of Apple, DELL, Intel, Lenovo, Microsoft, NVIDIA, Sony, Xiaomi and many more.

Please note that in-person conferences (i.e. AI Hardware Expo on May 5th-6th 2020, AI Enterprise Expo on August 25th-26th 2020 and AI Robotics Expo on November 12th-13th 2020 at SEMI, Milpitas, Silicon Valley) are converted into this virtual AI Expo series. 

Attendee Resources: Live Zoom Webinars, Community Rooms for each week/theme, Presentations and Recordings (after live sessions)