Learning Deep Learning with PyTorch Workshop Series - Institute for Digital Research and Education

This workshop series is to present overviews to the exciting deep learning techniques and to provide a practical guide for general audience to step into the field. It will be primarily appropriate for the beginners who want to learn the techniques and apply to their future research activities. Researchers with deep learning experiences are expected to get benefits from related discussions as well.
Please register for all 5 sessions. If you miss one session, there will be notes available to online for you to catch up for the next course.

Workshop 1: Introduction

In the first session of the series, we will give general introduction about machine learning, neural network and PyTorch. No specific prerequisite is required.

Workshop 2: Mechanics of Deep Learning

In the second session of the series, we will look into the procedures of working on a general deep learning project, especially on how to train a deep neural network. The knowledge of topics covered in the first session is assumed. Basic knowledge of Calculus and Linear Algebra will be helpful to understand the details.

Workshop 3: Using PyTorch

In the third session of the series, we will illustrate the basic usage of PyTorch and how to make deep learning project using PyTorch. Working experience of Python, Jupyter Notebooks will be helpful to follow the demos.

Workshop 4: Convolutional Neural Networks

In the fourth session of the series, we will introduce convolutional neural network and learning how to use PyTorch to do image processing for classic Dogs-vs-Cats problem. The knowledge of topics covered in the previous sessions is assumed. Working experience of Python, Jupyter Notebooks and linear algebra will be helpful.

Workshop 5: Transfer Learning

In the fifth session of the series, we will introduce transfer learning technology to use pre-trained models in PyTorch to get a better solution for deep learning projects. The knowledge of topics covered in the previous sessions is assumed. Working experience of Python, Jupyter Notebooks and linear algebra will be helpful.