Machine Learning (ML) Essentials: Deep-Learning & Multi-Modal Data Processing
By Mr. Sun Chung, Assistant Systems Librarian, HKU Libraries
Abstract
This seminar delivers insights through popular frameworks and architectures, deep learning fundamentals such as CNN and MNIST, and their applications in image processing. The session will introduce Multi-Modal AI by illustrating how models integrate diverse data sources to enhance performance in the Agentic era. The development of local multi-modal models is examined alongside using vector databases for efficient data retrieval.
Learning Outcomes:
After completing the session, learners will be able to
- Gain vision in deep-learning frameworks.
- Develop a clear understanding of multi-modal AI foundation, including model fine-tuning and integration of diverse data sources.
- Acquire knowledge on utilizing vector databases to enhance local multi-modal model deployments for efficient data retrieval.
Technical Focus Areas
- Deep learning fundamentals
- CNN architecture
- MNIST implementation
- Multi-modal AI overview
- Vector database concept
- Local model training
About the Speaker
Mr. Sun Chung
Assistant Systems Librarian, HKU Libraries
Mr. Sun Chung, serving as the Assistant Systems Librarian, is responsible for the administration and development support of library applications, integrated library systems such as Ex Libris Alma, and Machine Learning / AI frameworks in the library’s technology ecosystem. He works closely with other library units to explore and enhance library functionality and workflows.
Floor Plan of the Venue
