Undergraduate Seminar — Deep Learning with GPUs: The Next Frontier (Adam Thompson)
Deep Learning with GPUs: The Next Frontier
4:00 pm, September 28, 2022
326 St. Mary’s Hall
With the emergence of Deep Learning, GPUs (Graphics Processing Units) have become the accelerator of choice for speeding up a variety of software applications. To fully exploit the GPU’s potential, however, a programmer typically needs to learn low-level (often C/C++) software libraries, identify parallelism, and understand the GPU’s hardware architecture. Fortunately, thanks to recent developments with Python and associated open-source libraries, we can now gloss over some of the technical details and write GPU-enabled code via Python while stressing developer productivity and ease of use. In this talk, we’ll provide a blueprint for getting started with GPUs and even show some basic examples to cement core ideas. Hopefully, you’ll leave this talk with a firm idea of how GPUs work and how they can be used to make your software projects faster.
Adam Thompson is a Technical Product Manager at NVIDIA where he focuses on Edge HPC and Streaming Sensor solutions and is the creator of cuSignal – a GPU accelerated signal processing library written in Python with over 250,000 downloads. Adam’s technical interests involve signal processing, applications of deep learning to radio frequency data, high performance computing, and data compression. He holds a Masters degree in Electrical and Computer Engineering from Georgia Tech and a Bachelors Degree from Clemson University.