Abstract
Back in the day, machine learning problems were a drag. Either classification or data-generation, the main tasks of artificial intelligence, had been classified as computationally hard problems. Decades later, with subsequent computing innovation, machines have transformed into their ultra-smart, self-learning, automated versions that are sweeping the human landscape.
This unreasonable effectiveness of modern machine learning algorithms has thrown down the gauntlet to algorithms researchers, and there is perhaps no other problem domain with a more urgent need for the beyond worst-case approach.
In this talk, we will discuss a series of recent results in the area of Local Search, Continuous Optimization and Game Dynamics.