Anthill Inside 2017

On theory and concepts in Machine Learning, Deep Learning and Artificial Intelligence. Formerly Deep Learning Conf.


Getting Started with GPU Accelerated Deep Learning

Submitted by Gaurav Goswami (@gauravgoswami) on Monday, 10 July 2017

Section: Crisp talk Technical level: Beginner


Deep learning has been applied to various domains with great success and is a popular technique to solve challenging machine learning problems in the real world. However, deep learning is also computationally expensive and it is not feasible to train a deep network in a reasonable time frame on large databases without using GPU acceleration. In this talk, I will provide a tutorial on how to setup the pre-requisite drivers, packages, and tools to get started with GPU enabled deep learning on a machine running Ubuntu OS. In addition, we will take a quick look at running a deep learning workload on the GPU via a Jupyter notebook and conclude with an overview of the performance difference when using a CPU vs GPU for deep learning.


Draft slides are available here: I will update these to the full version soon. I would ideally like to showcase aspects of the workflow including the workload notebook live in action during the talk.

Speaker bio

I am currently working at IBM India as an Artificial Intelligence/Machine Learning expert. My Ph.D. thesis focuses on using machine learning and computer vision techniques to solve challenges in face recognition by improving the computation and combination of robust representations. I have had the opportunity to be a part of crafting a machine learning based solution for real world use cases in these domains.




  • Gaurav Goswami (@gauravgoswami) Proposer a year ago (edited a year ago)

    Preview Video:

    I have also added the preview video link to the slides.

    Slides have been updated to full version instead of just outline.

  • Gaurav Goswami (@gauravgoswami) Proposer a year ago

    Key takeaways:

    1. A quick overview of deep learning
    2. A step-by-step approach to setting up a GPU enabled environment.
    3. Step-by-step run through of a CNN based deep learning workload with steps that are common to a lot of problems
    4. A comparison of the performance of GPU vs CPU in training a deep learning architecture

Login with Twitter or Google to leave a comment