Search Results

This course will serve as a nexus between business analytics and cutting-edge artificial intelligence (AI) methodologies, which are increasingly vital in strategic business decision-making, encompassing areas from sales forecasting and inventory optimization to revenue management and supply chain analytics. There is a strong emphasis on the art, theory, and applications of Markov decision processes (MDPs) in modeling and optimizing large-scale business analytics problems in the presence of uncertainty. While MDPs can be applied to a broad range of decision analytics problems in practice, they often encounter the curse of dimensionality, which leads to exponential growth in the size of these models. To address this challenge, state-of-the-art AI methods, reinforcement learning (RL) and deep RL, are extensively discussed as solutions to overcome the curse of dimensionality. Additionally, if time permits, the course will present further recent advancements in AI-empowered decision analytics. This course will balance theory and practical applications of AI-empowered decision analytics, making it accessible to a diverse range of students and practitioners in fields like business, economics, finance, and engineering. Each lecture is composed of three components: theory, case studies, and tools. These are designed to equip students with an understanding of the underlying art and theory of MDPs and RL, insights into real-world applications, and proficiency in Python programming. Additionally, through a final group project, students will gain hands-on experience in the entire process of predictive and prescriptive analytics in practice. The final outcomes will be showcased through presentations.

Prerequisite(s): BU.510.601 AND BU.520.601 AND some skill and experience in programming required.