กิจกรรมประชาสัมพันธ์ : Boost MATLAB algorithms using NVIDIA GPUs

Invite instructor, researchers, staff and students of RMUTL to attend the seminar on the topic :  Deep Learning for Wireless Communications

Overview

This webinar is jointly presented by MathWorks and NVIDIA.

Researchers and engineers around the globe use MATLAB to analyze data, create algorithms or train models.
In this webinar, learn how MATLAB combined with NVIDIA solutions can accelerate and scale your work on GPUs.

At first, Axel (NVIDIA) presents the latest GPU options and features, including MIG (Multi-Instance GPUs), 3rd generation Tensor Cores, Mixed Precision and NVLink™. The focus will be on the NVIDIA A100 Tensor Core-GPU and how to leverage it in the cloud.

Building on that, Christoph (MathWorks) describes how you can take advantage of NVIDIA GPUs from within MATLAB without rewriting code for computationally intensive applications, for example in signal & image processing as well as deep learning. A case study is used to demonstrate how the usage of GPUs allows to scale computations. This is illustrated by the example of a parameter search for deep learning training, leveraging local and public cloud resources and comparing the performance.

Switching to the embedded area, Christoph shows how the power of NVIDIA embedded GPUs can be leveraged from MATLAB and Simulink via automated C++ and CUDA code generation, including a few example applications.

Robin (NVIDIA) then will further discuss NVIDIAs embedded GPU solutions, concluding the webinar with an update on the NVIDIA® Jetson™ platform and describes the hardware and software specifics, including NVIDIA JetPack SDK.

Highlights

We will cover the following topics:

• Speed up applications on NVIDIA GPUs without rewriting code
• Overview: latest NVIDIA hardware options on public clouds
• Case study on training the same neural network in various scenarios:
• CPU vs GPU vs multi-CPU vs. multi-GPU environments
• How to leverage the AWS and Azure clouds for flexible scaling of computations
• NVIDIA embedded GPU solutions
• Running deep learning inference on embedded GPUs through automated code generation

About the Presenters

Axel Koehler – Principal Solution Architect, NVIDIA
Axel Koehler has been with NVIDIA since January 2011. In his role Axel supports researchers, scientists, engineers and hardware and software partners in the implementation of GPU-based solutions for High Performance Computing (HPC) and Artificial Intelligence (AI). Prior to NVIDIA Axel worked at Sun Microsystems for 14 years in the global HPC team as lead architect and was responsible for the architecture and the design of large supercomputing installations.

Robin Roitsch – Embedded Business Development Manager EMEA, NVIDIA.
He holds an M.Eng. Degree in Software Engineering for Embedded Systems from the Technical University in Kaiserslautern. He started his role at NVIDIA in Mai 2020, focusing on the NVIDIA Jetson platform and the embedded market to support customers and partners to improve and scale business.

Christoph Stockhammer – Senior Application Engineer, MathWorks
He holds a M.Sc. degree in Mathematics from the Technical University Munich with an emphasis on optimization. He joined The MathWorks in 2012 and works as an application engineer. His focus areas include mathematics and data analytics, Machine & Deep Learning as well as the integration of MATLAB software components in other programming languages and environments.

For teachers, staff, staff 's RMUTL who are interested in attending the training Must proceed with registration through the system. to send information for attending the webinar according to the specified date and time

Register >>  Boost MATLAB algorithms using NVIDIA GPUs - MATLAB & Simulink (mathworks.com)