MIT

dols

Dynamics Optimization and Learning Systems

12.S592

  • Offered since 2017, in earlier form since 2013
  • Dynamics, Optimization, and Information foundations of Machine Learning
  • ML-based System Dynamics and Optimization for Earth, Planets, Climate, and Life
  • Flexible participation model
  • Pre-reqs: Linear Algebra, Probability & Stat, a first course in ML, and some Systems preparation (at least one of) Statistical Signal Processing, Estimation, Control or Optimization

Seminar course 12.S592 is offered every term. Its syllabus follows four primary themes annually with an almost periodic, ahem, “50-shades of challenging grey matter” over approximately 50 weeks.  Some have stayed on in the course for six years!

Theme By Season

  • We study the foundations of spaces and measures we use for Learning with Regularization and Sparsity. 

    1. Some Basics: Inner Products, Norms, Metrics, and Divergences
    2. Maximum Likelihood Estimation, Expectation Maximization & Mixtures
    3. Maximum A Posteriori Probability and Posterior Sampling
    4. Variational Inference, Conditional Entropy and the Information Bottleneck, Solving with Kernels
    5. Joint measures of “appearance” (amplitude) and “geometry” (phase).
      • Bayesian Coupled Optimal Transport and EstimationĀ