Deep Learning Jump-Start Workshop | SGInnovate
November 1-2
2018

Location

Perl @ BASH, Level 3,
79 Ayer Rajah Crescent
Singapore 139955

Deep Learning Jump-Start Workshop

Presented by SGInnovate and Red Dragon AI

SGInnovate partners with Red Dragon AI to introduce to you - Deep Learning Developer Series.

Please note that the dates for this workshop have not yet been confirmed and is subject to change.

This workshop is the second installation of the Deep Learning Series Workshop.  In this module, we go beyond the basic skills learned in module 1 such as Convolutional Neural Networks and expand your ability to build modern image networks using a variety of architectures and for applications beyond simple classification.

To understand the current state-of-the-art, we will review the history of ImageNet winning models and focus in on the Inception and Residual models.  We will also look at some of the cutting-edge models such as NASNet and AmoebaNet and show how they are different and how the field has gone beyond hand-engineered models.

One key skill that you will acquire is to learn how to use these modern architectures as feature extractors and then applying them to create applications like image search and similarity comparisons.

Start here

    • Have an interest in Deep Learning?
    • Join us if you are able to read and follow codes
    • This module is compulsory before you take the advanced modules
     
    • You will need to take module 1 before this module
     
    • You will need to take module 1 before this module
     
    • You will need to take module 1 AND module 2 OR 3 before this module
     
    • You will need to take module 1, 2 AND 3 before this module
    • Attain a “Deep Learning Specialist” certification when you complete all five modules
     
    • Have an interest in Deep Learning?
    • Join us if you are able to read and follow codes
    • This module is compulsory before you take the advanced modules
     
    • You will need to take module 1 before this module
     
    • You will need to take module 1 before this module
     
    • You will need to take module 1 AND module 2 OR 3 before this module
     
    • You will need to take module 1, 2 AND 3 before this module
    • Attain a “Deep Learning Specialist” certification when you complete all five modules
     

In this module you will also discover how to do such tasks as object detection and learn how models (like YOLO) are able to go beyond just classifying images to detecting where multiple objects are in an image.

You will also learn about image segmentation and classifying at the pixel level.  This will involve using architectures like U-Nets and DenseNets and learning how they are used in a variety of image segmentation tasks from perception for self-driving cars to medical image analysis.

Building on the tools taught in the first module we will be going beyond just using TensorFlow and Keras, to introduce PyTorch and TorchVision, which are often being used for research in computer vision and cutting-edge architectures.

As with all the modules you will have the opportunity to build multiple models yourself.  Most importantly, as part of your main project, you will be challenged to use your newly learned skills in an application that relates to your field of work or interest are.

Beyond giving you an understanding of what can be done in cutting-edge computer vision and how it is done, the goal of the workshop is to arm you deep learning computer vision skills so that you can apply it in your own area of work or project.

This workshop is pending funding approval. More details to be released soon. Please leave your contact details so that we can contact you the moment workshop is open for registration.

Workshop Overview:
In the course participants will learn:

  • Learn about advanced classification and objection detection
  • Introduction into PyTorch and TorchVision
  • Acquire skills to create applications like image search and similarity comparisons
  • Learn about image segmentation and classifying at the pixel level with architectures like U-Nets and DenseNets and how they are used in a variety of image segmentation tasks

The workshop will be held over 2 intensive days coupled with 6 hours’ worth of online content. This allows you to quickly learn the skills needed to apply Deep Learning and have access to ask your questions one on one. This is especially useful for understanding how to apply these skills to your unique applications.

Prerequisites:
Attended Module 1: Deep Learning Jump-start Workshop

Attendees MUST bring their own laptops

Agenda:

Day 1
Section 1: Convolutional Neural Network Recap
Topics Covered

  • Convolution Math in layers
  • Pooling and Strides
  • Alexnets
  • Building CNN networks
  • Calculating the parameters and shapes of various networks
  • Tuning CNN
  • VGG Network

Section 2: Intermediate CNNs
Topics Covered

  • Modern Convolutional Nets
  • Transfer Learning with CNNs and Finetuning
  • Inception architectures
  • Residual Networks
  • Imagenet history and applications
  • Building a classifier using transfer learning
  • Kaggle competition for images part 1
  • Start personal project 1

Day 2
Section 3:
Topics Covered

  • Auto Encoders
  • Repurposing CNN models
  • Object Detection 
  • YOLO
  • Build an Image search system
  • Continue personal project 1

Section 4: CNNs Segmentation
Abstract:

  • Image Search
  • Segmentation Networks
  • U-Net and Skip connections architectures
  • Batch Normalization

Section 5: Video Walk throughs
Topics Covered:

  • Building CNNs from scratch
  • Building Auto Encoders
  • Understanding Object detection and location models
  • Style Transfer
  • Fast Style Transfer

Instructors’ Biodata:
Dr Martin Andrews
Martin has over 20 years’ experience in Machine Learning and has used it to solve problems in financial modelling and has created AI automation for companies. His current area of focus and specialty is in natural language processing and understanding. In 2017, Google appointed Martin as one of the first 12 Google Developer Experts for Machine Learning.

Sam Witteveen
Sam has used Machine Learning and Deep Learning in building multiple tech start-ups, including a children’s educational app provider which has over 4 million users worldwide. His current focus is AI for conversational agents to allow humans to interact easier and faster with computers. In 2017, Google appointed Sam as one of the first 12 Google Developer Experts for Machine Learning in the world.

Topics: Artificial Intelligence / Deep Learning / Machine Learning / Robotics

You may also like the following: