Presented By O’Reilly and Intel Nervana
Put AI to work
September 17-18, 2017: Training
September 18-20, 2017: Tutorials & Conference
San Francisco, CA

In-Person Training
NVIDIA Deep Learning Institute bootcamp

Mike Mendelson (NVIDIA)
Sunday, September 17 & Monday, September 18, 9:00am - 5:00pm
Location: Franciscan B
Secondary topics:  Deep learning
Average rating: *****
(5.00, 1 rating)

Participants should plan to attend both days of this 2-day training course. Platinum and Training passes do not include access to tutorials on Monday.

NVIDIA Deep Learning Institute-certified instructor Mike Mendelson walks you through solving the most challenging problems with deep learning. You'll start with deep learning basic concepts and quickly move to taking on real-word problems using deep learning.

What you'll learn, and how you can apply it

  • Understand deep learning basic concepts and terminology
  • Explore the fundamentals of deep learning to train neural networks and use results to improve performance and capabilities
  • Learn how convolutional and recurrent neural networks can be combined to generate effective descriptions of content within images and video clips

Hardware and/or installation requirements:

  • A laptop with Chrome, Firefox, or Safari installed (Internet Explorer is operational but does not provide the best performance.)

NVIDIA Deep Learning Institute-certified instructor Mike Mendelson walks you through solving the most challenging problems with deep learning. You’ll start with deep learning basic concepts and quickly move to taking on real-word problems using deep learning. Upon completion of the two-day workshop, you’ll receive an NVIDIA Deep Learning Institute certificate of attendance.

Outline

Day 1

9:00am–9:30am: Deep learning demystified and applied deep learning

  • Lecture: General background on deep learning, key terminology, use cases from various industries, how deep learning differs from previous algorithmic approach, and how a deep neural network gets trained, optimized, and deployed. Plus, learn how to apply deep learning to challenging problems, what types of problems benefit most from deep learning, what skills and knowledge are needed to use deep learning, and the characteristics of successful deep learning projects.

9:30am–10:30am: Image classification with DIGITS

  • Hands-on exercise: Deep learning is giving machines near human levels of visual recognition capabilities and disrupting many applications by replacing hand-coded software with predictive models learned directly from data. This lab introduces the machine learning workflow and provides hands-on experience with using deep neural networks (DNN) to solve a real-world image classification problem. You’ll walk through the process of data preparation, model definition, model training and troubleshooting, validation testing, and strategies for improving model performance. You’ll also see the benefits of GPU acceleration in the model training process. By the end, you’ll be able to use NVIDIA DIGITS to train a DNN on your own image classification dataset.

10:30am–11:00am: Morning break
 
11:00am–12:30pm: Image classification with DIGITS continued

12:30pm–1:30pm: Lunch
 
1:30pm–3:00pm: Object detection with DIGITS

  • Hands-on exercise: This hands-on lab gives you tools for solving unsolved problems in deep learning. You’ll start with existing solutions and combine deep learning models with traditional programming, modify the internal math (layers) of a neural network, and identify and make use of solutions to similar problems.

3:00pm–3:30pm: Afternoon break
 
3:30pm–5:00pm: Neural network deployment with DIGITS and TensorRT

  • Hands-on exercise: Whether you’re classifying images to help fight cancer or detecting objects to avoid pedestrians in self-driving cars, deployment is where you’ll solve a problem with deep learning. Your network uses what it learned during training to make decisions in the real world. You’ll use three different interfaces to pass new data through a trained network, optimize networks for performance during inference, and learn to deploy networks that they have and have not trained.


Day 2

9:00am–10:30am: Image segmentation with TensorFlow

  • Hands-on exercise: Image (or semantic) segmentation is the task of placing each pixel of an image into a specific class. You’ll classify each pixel in a cardiac MRI image based on whether the pixel is a part of the left ventricle (LV) or not. You’ll compare image segmentation with other computer vision problems, experiment with TensorFlow tools such as TensorBoard and the TensorFlow Python API, and learn metrics for assessing segmentation performance.

10:30am–11:00am: Morning break

11:00am–12:30pm: Word generation with TensorFlow

  • Lecture and exercises: Introduction to natural language processing. You’ll use a recurrent neural network (RNN) to predict the next word of a sentence. You’ll learn how to preprocess textual data using one-hot encoding, how factors like dictionary size, network depth, and other parameters affect performance, and how to load data and train and test RNNs.

12:30pm–1:30pm: Lunch
 
1:30pm–3:00pm: Image and video captioning by combining CNNs and RNNs

  • Lecture and exercises: In order to solve large problems with deep learning, one type of network may not be enough. You’ll combine the output of a convolutional neural network (CNN) with the input of a recurrent neural network to generate descriptions of scenes. You’ll examine the output of internal layers of CNNS, combine networks by concatenating vectors, and interpret context in video data by using what you know about image features.

3:00pm–3:30pm: Afternoon break
 
3:30pm–5:00pm: Image and video captioning by combining CNNs and RNNs continued | Wrap-up and Q&A

About your instructor

Michael (Mike) Mendelson is a curriculum designer and certified instructor at NVIDIA’s Deep Learning Institute. Mike began experimenting with deep learning while working to enable active personalized (human) learning. Previously, he built and taught world-class project-based STEM curriculum at EL Education. Mike is inspired by the power of deep learning to solve some of the world’s most important challenges.

Conference registration

Get the Platinum pass or the Training pass to add this course to your package.

Comments on this page are now closed.

Comments

Picture of Jason Perdue
Jason Perdue | SPEAKER MANAGER
09/11/2017 1:37am PDT

Here is the venue information:

https://conferences.oreilly.com/artificial-intelligence/ai-ca/public/content/hotel

The room is listed above: Franciscan B

Jignesh Mehta | DIRECTOR, BIG DATA AND ANALYTICS
09/10/2017 4:29pm PDT

Would you please let us know exact address/location, room/conference area for this session? It seems I am not able to find it anywhere. Thanks!

Allison Rhea Jones | CONFERENCE CONTENT MANAGER
08/24/2017 3:55am PDT

@Hengliang Tian You do need to bring your own laptop. Attendees will access NVIDIA GPUs on Amazon Web Services so all you need is a laptop with a current browser.

Hengliang Tian | STATISTICAL RESEARCH MODELER
07/20/2017 6:56am PDT

do we need to bring our own laptop? I think NVDIA graphic card is required for this class. Right?