Skip to content
0:00 / 0:00

#graphcore#aiaccelerator#ipu

RAAIS 2017 - Simon Knowles, Co-Founder and CTO of Graphcore

The Research and Applied AI Summit - RAAIS

4.1K subscribers • 1 video on Hallway

YouTube
  • Recommended by 3 people: 3 on Hacker News.
  • Presented by Simon Knowles, co-founder and CTO of Graphcore
  • Made by the vendor. The company whose product it covers made or sponsored it.
  • May be out of date. Uploaded 9 years ago, in an area where tools change fast.

Advanced. Deep dive for people who know the field.

6.8K views9 years ago3 recommendations

HardwareAI and machine learningCPUs and computer architectureNeural networks and deep learning

Simon Knowles explains the design of intelligence processing units and how moving memory closer to processing cores accelerates machine learning workloads.

Simon Knowles on "Intelligence Processors". Simon is co-founder and CTO of Graphcore. He co-founded and exited two fabless semiconductor companies, Element14 and Icera, for a combined value of over $1billion. Before that he headed the microprocessor development group at ST Micro. He holds an MA in Electrical Science from Cambridge University and is the author of 14 issued patents. graphcore.ai __________ RAAIS is a community for entrepreneurs and researchers who accelerate the science and applications of AI technology for the greater good. The London based AI conference is a one-day event covering machine learning, deep learning, reinforcement learning, data science, healthcare and life sciences. It's presented by The RAAIS Foundation, a non-for-profit with a mission to increase participation, access to education and resources in AI in order to help advance the greater good. RAAIS: raais.co/ The RAAIS Foundation: raais.org/ Air Street Capital (airstreet.com/) is a venture capital firm investing in AI-first technology and life science companies.

Why it's here

3 people recommend this

What people say. Recommended for its technical explanation of AI hardware design and the benefits of collocating memory and processing cores.

visargaon Hacker News • Recommends

On “Neuromorphic Computing with Nanoscale Spintronic Oscillators”

Interesting approach, but very far off. I prefer GraphCore as a promising new lead in AI hardware. Moving computation close in with memory, and making data transfers async will speed up DL. link to this video

midkoon Hacker News • Recommends

On “Intel to Develop Discrete GPUs, Hires AMD's Raja Koduri as Chief Architect”

Indeed, and that's why there are a couple of startups working on new chips and why Google has the TPU. Here's a nice technical talk from Graphcore's CTO about that link to this video

maffydubon Hacker News • Recommends

On “The Looming Battle Over AI Chips”

…to your car starter motor!) and also getting the heat out. Logic gates take a lot more power than RAM, so he argued that collocating small chunks of RAM right next to your processing core was much better from a power perspective (meaning you could then pack yet more into your chip) as well as obviously being better from a performance perspective. link to this video isn't quite the presentation I saw, but it has quite a lot of overlap. Hope that helps!…