'IT Support for Research' Workshop: Deep Learning with Python (Afternoon class)

Deep learning is a subfield of machine learning based on the application of artificial neural networks which are comprised of artificial neurons that stimulate biological neurons. The use of multiple layers of neurons allows a model to learn much more complicated features compared with classical machine learning algorithms, and therefore it is widely used in difficult tasks such as computer vision and speech recognition.

The workshop, consists of five online sessions, will focus on working with an open source AI library named Tensorflow, developed by Google Brain, to build neural networks that handle image recognition, speech synthesis and advanced time series prediction. The workshop will also cover various optimization algorithms and regularization techniques for fine tuning a neural network that utilizes model performance.

Students have basic programming knowledge in Python are welcome to join.

'IT Support for Research' Workshop: Deep Learning with Python

(Afternoon class)

Date: 27 Jan (Thu), 10 Feb (Thu), 17 Feb (Thu), 24 Feb (Thu), 3 Mar (Thu)

Time: 14:30 – 17:00

Target Audience: RPg, TPg and Ug students

Pre-requisite: Basic programming knowledge in Python 

Medium of Instruction: English

Course Outline:

Lesson 1

1. Deep Learning with Tensorflow
   1. Tensor and Operations
   2. Linear Regression
   3. Deep Neural Network

Lesson 2

1. Convolutional Neural Network (CNN)
   1. CNN on Image Classification
   2. Augmentation
   3. Transfer Learning

Lesson 3

1. Natural Language Processing
   1. Tokenization and Word Embedding
   2. Transformer and BERT

Lesson 4

1. Time Series Forecasting
   1. RNN and LSTM
   2. Temporal Fusion Transformer

Lesson 5

1. Neural Network Hyperparameter Tuning
   1. Keras Tuner
   2. HParams