Deep Learning and Applications, CMPT 498/898, offered in Winter 2019/2020

Dr. Ian Stavness will be teaching a special interest class, Deep Learning and Applications, in the upcoming term.

This class will be a survey of Deep Learning techniques and their application to problems in computer vision and data science. Deep learning techniques may include Deep Neural Networks, Convolutional Neural Networks, Recurrent Networks, Deep Generative Models and Reinforcement Learning. Application domains will focus on computer vision problems, including image classification, object detection, and image segmentation. Additional application domains in natural language processing and robotics control will be introduced. Software tools will be introduced for practical application.

Prerequisites for this class are: (Math 164 or 266, or EE 215, or CE 318), STAT 245, and (CMPT 317 or CMPT 487). Prerequisites may be waived by written instructor approval.

Please email instructor Ian Stavness for approval: ian.stavness@usask.ca

 

Term: Winter 2019/2020

Day: Tuesday/Thursday (TR)

Time: 2:30pm - 3:50pm

Location: TBA

 

Learning Outcomes:

This course provides students with a survey of deep-learning techniques applied to computer vision and other data science problems. By the completion of this course, students will be expected to:

  • Understand the underlying mechanisms and differences among the main classes of deep learning architectures, including deep neural networks, convolutional neural networks, recurrent neural networks, generative models and reinforcement learning;
  • Know which of the deep learning architectures are best suited for different computer vision tasks, including image classification, image segmentation, instance segmentation, and regression problems involving image data;
  • Understand generally how deep learning is in used other application domains, including natural language processing, robot control, and computer graphics;
  • Be sufficiently knowledgeable to read and synthesize material from recent research papers that apply deep learning techniques to computer vision problems; and
  • Be proficient at developing software for training and testing a deep learning models for computer vision tasks, such as image classification and object detection.