About : This course will focus on building various multiagent models with strategic agents and then study algorithms for decision making in those settings. After going over some basics of game theory we will cover topics such as network games, equilibria computation through online learning, Stochastic games and Infinite games. We will also see how this theory is applied to modern ML paradigms such as Generative adversarial networks (GANs), multiagent Deep RL, debate framework in LLMs and other large scale multiplayer games such as DeepMind's AI agent for Starcraft.
Course Prerequisite : A basic course on game theory is preferred, but not necessary as the required concepts will be covered. Students are expected to have sufficient mathematical maturity, such as having working knowledge of linear algebra, multivariable calculus, (undergraduate) probability and ability to understand and write proofs.
Level : Advanced Undergraduate, Masters or PhD.
Credits: This course has a graded oral exam at the end of the semester and the students who pass the exam will obtain 5ECTS.
Detailed Lecture Schedule : TBC
Location and Timing : TBC
- Trainer/in: Sai Ganesh Nagarajan