Teaching
Modules for 2025-26
Advanced Artificial Intelligence
Module code: COMP219
Role: Module Co-ordinator
Supervised Theses
- Advancing Geometric Deep Learning: Deep Architectures and Latent Geometry
- Adversarial Robustness of Graph Neural Networks
- Analysis for Data with Uncertainty based on Fuzzy Theory
- Enhancing Autonomous Vehicle Perception: A Focus on Embedding Learning and Uncertainty Analysis
- Exploring Deep Learning Approaches for Solving Plane Geometry Problems
- Investigating and Improving Real-World Representation Learning through Causal Inference
- Learning Graph Neural Networks on Complex Structured Data
- Learning to Utilize Potential Inner Relationships Toward 3D Point Cloud Completion
- Long-Tailed Oracle Character Recognition
- Matching and Segmentation for Multimedia Data
- Person Re-identification with and without Supervision
- Reliable Deep Neural Networks with Randomised Weights
- Stealthy attacks and Ensemble defence in the Neural Network arena
- Study of Robust and Interpretable Zero-shot Image classification
- Towards Exploring and Enhancing the Robustness of Deep Neural Networks
- Towards Inference Efficient Spiking Neural Network
- Towards Reliable Deep Learning Models: Risk Quantification and Robustness Verification
- Towards Reliable Deep Learning Models: Risk Quantification and Robustness Verification
- Towards Safe and Robust AI Agents: Data Augmentation, Model Distillation, and Inference-Time Alignment
- Verification and Validation of Machine Learning Safety in Learning-Enabled Autonomous Systems