TalentQ Seminar – Marco Cerezo

Speaker: Marco Cerezo

Title: Is Quantum Machine Learning an ill-defined framework?

Abstract: In this talk we will discuss some our recent results regarding the trainability and classical simulability of Quantum Machine Learning (QML). Despite the initial hype, it has been shown that many QML models can exhibit critical issues such as barren plateaus, and exceeding local minima in their training landscapes. More importantly, it has been recently pointed out that models which are barren plateau-free, could also be potentially classically simulated. As such, we will discuss the possibility that the QML framework could be ill-defined.

Registration: https://events.teams.microsoft.com/event/3e5cd889-69ba-407e-a01d-348e64e2def7@8f0d452c-b7a4-4964-b810-8c397374477b/registration