Introduction to Deep Learning
I was head TA and organizer for the Introduction to Deep Learning course at the University of Washington in Fall 2019, teaching the basics of deep learning to more than 160 students.
I wrote the homeworks, including an assignment in which students implement layers in a neural network library I created using only Python, NumPy and Numba. We also used Colab notebooks and PyTorch to train real models on GPUs.
I taught alongside Ali Farhadi, who ran the course, and nine other TAs. The slides below are the ones used in the course. If you are an instructor and would like to reuse the course materials, get in touch.
Slides
- Machine Learning Review
- Neural Networks and Optimization
- Neural Networks in Practice
- Convolutional Neural Networks
- Image Classification
- Network Architectures
- Segmentation and Detection
- Dealing with Sequential Data
- BERT / Transformers
- Optimizers
- Making Deep Learning Work
- Deep Reinforcement Learning
- Generative Adversarial Networks