Deep Learning - Graduate Course (Teaching Assistant)

Continued as Teaching Assistant for Prof. Stan Z. Li’s Deep Learning course in the second semester. Building on the experience from the previous year, enhanced the course with updated content on recent advances and more challenging project assignments.

Responsibilities

• Curriculum Development: Helped update course content with latest research advances including Vision Transformers, Large Language Models, and Diffusion Models
• Advanced Tutorials: Designed and delivered advanced tutorials on distributed training, model optimization, and efficient deployment
• Project Supervision: Supervised more complex research-oriented projects connecting to real-world applications
• Guest Lectures: Delivered guest lectures on manifold learning and dimensionality reduction techniques
• Research Integration: Connected course topics with ongoing research in AI for Science applications

Enhanced Topics

• Vision Transformers (ViT) and Swin Transformers
• Pre-training and foundation models
• Self-supervised learning (contrastive learning, masked modeling)
• Diffusion models for generative AI
• Neural architecture search
• Efficient deep learning (quantization, pruning, knowledge distillation)
• Geometric deep learning and graph neural networks
• Applications in drug discovery and biological data analysis

Teaching Innovation

Introduced project-based learning with industry collaboration, where students worked on real datasets from biotechnology and medical imaging companies. Organized a mini-symposium where students presented their final projects to invited researchers and industry partners, providing valuable feedback and networking opportunities.

Student Outcomes

Several student projects led to conference paper submissions and two students were successfully recruited as research interns in leading AI research labs based on their project work.