🎓 Undergraduate Student
🏫 Queen Mary University of London (QMUL)
🎓 Incoming Master's Student
🏫 University of Waterloo (Starting Sept 2026)
My research explores how to build adaptive, reliable, and self-evolving machine learning systems.
I began my academic journey focusing on deep learning for computer vision, particularly:
- Generative Models (Diffusion Models, GANs)
- Medical Image Analysis
- Knowledge Distillation
- Representation Learning
Through this work, I became increasingly interested not only in how models perform, but in how they adapt, reason, and improve over time.
My current research direction shifts toward foundational questions in machine learning systems.
I am particularly interested in:
- Trustworthy Machine Learning Systems
- Self-Adaptive Model Architectures
- Machine Metacognition
- Continual Learning
- Parameter-Efficient Fine-Tuning (PEFT)
Broadly, I aim to design models that:
- Understand their own uncertainty
- Adapt to distribution shifts
- Learn continuously without catastrophic forgetting
- Scale efficiently without full retraining
Core Areas
- Deep Learning Theory & Practice
- Generative Modeling
- Efficient Model Design
- Medical AI Systems
Tools
- Python
- PyTorch
- C++
- LaTeX
- Diffusion-based MRI Enhancement
- Mamba-based Medical Image Segmentation
- Targeted Knowledge Distillation
- Adaptive and Efficient Model Architectures
I am interested in pursuing research at the intersection of:
Adaptive Intelligence × Trustworthy Systems × Scalable Learning
with the long-term goal of contributing to next-generation self-improving machine learning systems.
- 📧 Email: thisisjamesi040323 at G mail dot com