Introduction
Deep learning is a class of machine learning algorithms which enables computers to learn from examples. Deep learning techniques have been used successfully for variety of applications, including: automatic speech recognition, image recognition, natural language processing, drug discovery, and recommendation systems etc. This course will provide students to learn the fundamentals of deep learning, the architecture and core components of neural networks, NN optimization, hyperparameter selection and how to implement, train, and validate their own neural network (mainly an application of images).
The reuse of a previously learned model on a new problem is known as transfer learning. It is particularly popular in deep learning since it can train deep neural networks with a small amount of data and produce an accurate model. The knowledge of an already trained model for a long time on huge datasets is transferred to a different but closely linked problem throughout transfer learning. In this course, students will learn the evolution of transfer learning, Transformer architecture and how to implement, train, pre-train transformer-based models on a selected use case (mainly an application of NLP).
Organisation
14 hours coursework, 12 hours practice
Evaluation
Practical session grading
Posted 28 September 2023