Data Science Talk Series 

Title: Complex-Structured Optimization Problems in Machine Learning

Abstract: In recent years, complex-structured optimization problems have become fundamental challenges in multi-agent and multi-task learning applications, including recommendation systems, autonomous drones, and cybersecurity. This talk will delve into these challenges, focusing on min-max optimization, bilevel optimization, decentralized learning, and federated learning—key paradigms that shape modern machine learning. By bridging theoretical insights with practical applications, this talk will provide a deeper understanding of how advanced optimization techniques enhance robustness, efficiency, and security in machine learning systems.