Academics

  /  

Courses

  /  

Descriptions

  /  

Keep

IEMS 490: Optimization and Learning in Stochastic Dynamic Environments

VIEW ALL COURSE TIMES AND SESSIONS

Prerequisites

probability theory, real analysis or equivalent for mathematical maturity.

Description

Decision making in stochastic and dynamic environments plays an essential role in many areas, including finance, robotics, game theory, revenue management and social networks. This course aims to gain a theoretical understanding for some popular tools used to solve these problems.  Topics include: theoretic foundations of reinforcement learning/dynamic programming, multi-armed bandit problems and its solutions, random graph models for social and economic networks which can be used to explain small world phenomena, wealth concentration and disease/information spreading.

Grades:
Problem Sets – 25%
Midterm – 35%
Final Project – 40%