Welcome to the Fall 2024 offering of CSE 6243!

Prerequisite: Graduate level Machine Learning, Probability and Statistics, Numerical Linear Algebra

This class is best for you if you have machine learning at the CORE of your studies/research, and want to understand the fundamentals.

Advanced Machine Learning is a graduate level course introducing the modern machine learning techniques, including graphical models, optimization, generative models, reinforcement learning, etc. These topics will be covered in Four Modules:

  • Module I: Background Knowledge
    • Convex Optimization: Convex function, duality, stochastic gradient descent
    • PGM: directed/undirected graphical models
    • Sampling: MH, Gibbs, HMC
  • Module II: Generative Model
    • VAE, Autoregressive Model, GAN, EBM (CD, Score Matching), Diffusion Models
  • Module III: Representation Learning
    • Contrastive and Non-Contrastive Representation Learning, Multi-Modality Representation Learning.
    • Representation Learning vs. Generative Models.
  • Module IV: Reinforcement Learning
    • MDP, Dynamic Programming, Policy-gradient, Imitation Learning
    • (Offline RL, Exploration)

The course will be very math and theory heavy. The students are expected to develop the skill of design computationally and statistically efficient machine learning algorithms for practical problems.

Time: Monday/Wednesday 5:00-06:15 pm
Location: Molecular Sciences and Engr 1222
Discussion & HW submission: Canvas and Ed Discussion
Contact: Please ask all course-related questions on Ed Discussion, where you will also find announcements. For external enquiries, personal matters, or in emergencies, you can email us at bodai at cc.gatech.edu and shribak at gatech.edu.