 |
Center for Nonlinear Studies |
In this talk, I will discuss how, by drawing inspiration from foundational questions (such as "Where does time emerge from?") standard notions of quantum correlation can be extended to new domains, thus enabling the development of innovative quantum, classical, and hybrid schemes for simulating quantum many-body physics. As a first example, I will introduce the concept of "parallel-in-time" computation, a quantum algorithmic framework that exploits the recent concept of system-time entanglement to efficiently estimate large temporal averages of physical quantities. As a demonstration, I will describe applications to the problem of equilibration of a many-body quantum system exhibiting Anderson localization. As a second example, I will describe how a sensible notion of fermionic entanglement, able to capture their intrinsic indistinguishability, determines the trainability of certain families of quantum machine learning algorithms. I will also comment on how these two examples can be combined in hybrid classical-quantum schemes that might not require full scale quantum computing to be implemented. Finally, I will outline open problems and offer additional insights that extend beyond these initial examples.
Microsoft Teams: Join the meeting now
Meeting ID: 251 441 470 950
Passcode: ry2NY3SH
Host: Marco Cerezo de la Roca (CCS-3)