Stanford Seminar: Deep Learning in the Age of Zen, Vega, and Beyond

EE380: Computer Systems Colloquium Seminar
Computer Architecture : Deep Learning in the Age of Zen, Vega, and Beyond
Speaker: Allen Rush, Advanced Micro Devices, Inc.

Deep Learning and Machine Intelligence is maturing to the point where is it is being deployed to many applications, particularly large data, imaging classification and detection. This talk addresses the challenges of deep learning from a computational challenge perspective and discusses the ways in which new compute platforms of Zen (x86) and Vega (GPU) provide high performance solutions to different training and inference applications. The ROCm software stack completes the support with libraries and framework support for a variety of environments.

About the Speaker:
Allen Rush is a fellow at AMD, focusing on imaging and machine learning architecture development. He has been active in imaging and computer vision projects for over 25 years, including several startups. He is the domain architect for ISP and current machine learning development activities in HW, SW and application support.

Support for the Stanford Colloquium on Computer Systems Seminar Series provided by the Stanford Computer Forum.

Colloquium on Computer Systems Seminar Series (EE380) presents the current research in design, implementation, analysis, and use of computer systems. Topics range from integrated circuits to operating systems and programming languages. It is free and open to the public, with new lectures each week.

[akompsupport]

Thanks for posting. First time I have heard of AMD's Open Compute initiative. I think it's great. And they can re-compile CUDA and CUNN to run on AMD GPUs? Well worth watching.

[giannagiavelli]

Our first product is a 2, 000 level network with varying architectures, 8 billion neural units.

[giannagiavelli]

Hinton is what i studied in the 1980s, today he has made no progress. Neural cube designs are far more interesting and will change the world. Based on Edelmen and Pribram's work, they are the state of the art. Holographic recognition planes, Ontic Processors, and Giav rated mega-billion neuron systems are our current designs at Noonean, we integrate vision, active learning, thought, language and memory, not Googelian 256x256 image frames. We are interested in the AMD MI25 instinct series but unfortunately the card design is quite gigantic for embedded systems.