Learning Machines
Key questions
- 375:
- How can systems improve from experience?
- What can be learned from data vs interaction?
- How can we evaluate learning: does it generalize?
- 376:
- How can we learn without labeled data? (self-supervised learning)
- How do foundation models learn generalizable patterns from massive datasets?
- How can generative agents learn to improve their behavior from feedback?
Key objectives
After this course, I will be able to:
- 375:
- Learning Theory
- I can explain how different ML approaches (supervised, unsupervised/self-supervised, reinforcement) learn.
- I can explain how stochastic gradient descent uses data to improve performance.
- I can describe the relationship between loss functions and metrics.
- Implementation and Debugging
- I apply data validation techniques without having to be reminded to do so.
- I can diagnose problems in model training, such as overfitting or underfitting, from metrics.
- I can implement a basic training loop in PyTorch (also in ML Engineering)
- 376:
- I can explain how next-token prediction implicitly encodes many other linguistic and reasoning tasks.
- I can explain how supervised tuning and feedback tuning can improve the performance and reliability of a model.
Materials
Contents
Classification Models (draft!)
The content may not be revised for this year. If you really want to see it, click the link above.
Generalization and a Kaggle Competition (draft!)
The content may not be revised for this year. If you really want to see it, click the link above.
Regression Models (draft!)
The content may not be revised for this year. If you really want to see it, click the link above.