세미나

📅 Seminar Information

🔸Date: April 1, 2025 (Tue) 16:00

🔸Location: Building 104, Room E104

🔸Speaker: Ph.D Hyunglip Bae (KAIST)

🔸Topic: Enhancing Stochastic Optimization Algorithms through the Application of Deep Learning Models

– This seminar will be conducted in English

📑Abstract: 

In this talk, I will introduce approaches to improving stochastic optimization algorithms by applying deep learning models. Traditional optimization methodologies require re-running the entire algorithm from scratch, even for similar parameter settings, without utilizing previous optimal solutions. To address this limitation, I will present a novel algorithm for solving a single multistage stochastic programming problem and a model for approximating multiple multistage stochastic programming problems. Additionally, I will discuss the recently emerging concept of decision-focused learning, which aims to predict uncertain parameters while considering the underlying optimization problem. Finally, I will share insights into my ongoing research and future research directions.

👤Pesenter Bio: 

Hyunglip Bae received his Ph.D. in Industrial and Systems Engineering from the Korea Advanced Institute of Science and Technology (KAIST) in February 2025. His dissertation, titled “Study on Approximation of Large-Scale Parametric Optimization via Deep Learning with Applications on Financial Planning,” was supervised by Professor Woo Chang Kim. During his doctoral studies, he authored three papers on the topic of AI for Optimization at leading AI conferences. Currently, he is a postdoctoral researcher at KAIST, where he continues his research in Financial Engineering, Stochastic Programming, and Decision-Focused Learning.