The aim of this module is to provide students, who have foundational knowledge of machine learning and deep learning, with a broad understanding of artificial intelligence and large-scale machine learning models, including their theoretical underpinnings, development, and practical applications. It guides students to design, train, and implement large-scale models. It also aims to offer hands-on experience in applying large-scale models to real-world problems, particularly in natural language processing (NLP) and computer vision (CV), ensuring that students can bridge the gap between theory and practice.
A Demonstrate theoretical and practical knowledge of large-scale machine learning model technologies. B Apply and adapt pre-trained large-scale machine learning models for specific tasks. C Analyze and critically assess the deployment and operational impact of large-scale machine learning models in industry-relevant examples. D Evaluate machine learning algorithms and models for industrially relevant problems.
This module employs a blended learning strategy to develop students' theoretical understanding and practical skills in large language model applications. Interactive lectures introduce foundational concepts, including transformer architectures, pre-training paradigms, reinforcement learning from human feedback, and application techniques such as prompt engineering, tool augmentation, and retrieval-augmented generation, guiding students from core principles to industrial implementation. Practical lab sessions provide hands-on experience with industry-standard tools and datasets, enabling students to experiment with fine-tuning frameworks, databases, and API integration. Core technical skills required for the coursework are systematically exercised across these sessions, with one dedicated lab addressing advanced implementation challenges specific to the assessment. Students receive individual academic support through regular office hours and are encouraged to take initiative by reviewing research literature, exploring software tools, and advancing their project work independently. All course content and supplementary materials are accessible through the University's Learning Mall platform to facilitate continuous and self-directed learning.