Course material & Lecture notes

■ Nonlinear Optimization (MIT 6.7220 / 15.084, Spring 2024)

Introduction to the fundamentals of nonlinear optimization theory and algorithms. When applicable, emphasis is put on modern applications, especially within machine learning and its sub-branches, including online learning, computational decision-making, and nonconvex applications in deep learning.

Course material
2024-02-06
html | pdf
2024-02-08
html | pdf
2024-02-13
html | pdf
2024-02-15
html | pdf
2024-02-22
html | pdf
2024-02-27
html | pdf
2024-02-29
html | pdf
2024-03-05
html | pdf
2024-03-07
html | pdf
2024-03-12
html | pdf
2024-03-14
html | pdf

■ Topics in Multiagent Learning (MIT 6.S890, Fall 2023)

This new graduate course, co-developed with Costis Daskalakis, presents the foundations of multi-agent systems from a combined game-theoretic, optimization and learning-theoretic perspective, building from matrix games (such as rock-paper-scissors) to stochastic games, imperfect information games, and games with non-concave utilities. We present manifestations of these models in machine learning applications, from solving Go to multi-agent reinforcement learning, adversarial learning and broader multi-agent deep learning applications. We discuss aspects of equilibrium computation and learning as well as the computational complexity of equilibria. We also discuss how the different models and methods have allowed several recent breakthroughs in AI, including human- and superhuman-level agents for established games such as Go, Poker, Diplomacy, and Stratego.

Course material
2023-09-19
html | pdf
2023-09-21
html | pdf
2023-10-24
html | slides
2023-10-26
html | slides
2023-10-31
html | slides
2023-11-02
html | slides
2023-11-07
html | slides
2023-11-09
html | slides

Course Homepage

■ Computational Game Solving (CMU 15-888, Fall 2021)

This new graduate course, co-developed with Tuomas Sandholm at CMU, focuses on multi-step imperfect-information games. Imperfect-information games are significantly more complex than perfect-information games like chess and Go, and see emergence of signaling and deception at equilibrium. There has been tremendous progress in the AI community on solving such games since around 2003. The course covers the fundamentals and the state of the art of solving such games.

Course material
2021-09-09
html | pdf
2021-09-14
html | pdf
2021-09-16
html | pdf
2021-09-21
html | pdf
2021-09-28
html | pdf
2021-09-30
html | pdf
2021-10-05
html | pdf
2021-10-07
html | pdf
2021-11-11
html | pdf
2021-11-16
html | pdf

Course Homepage

Reports of typos are always welcome! Please reach out at gfarina AT mit.edu.