Informationally complete generalised quantum measurements for near-term quantum algorithms

Speaker: Matteo Rossi, Algorithmiq

Time & Date: Wedneyday, November 30th, 15.00

Hosts: Philip Walther & Borivoje Dakic

Abstract:
Today’s quantum computers are imperfect. They are made of dozens of qubits that can be prepared in highly non-classical states but, being very sensitive to noise, their ability to preserve quantum properties is very limited. This is why, despite the discovery of algorithms that in principle would allow us to simulate interesting and currently intractable problems in chemistry and materials, applications to relevant problems seem out-of-reach. In this talk I will present recently-introduced techniques that allow near-term quantum computers to run quantum algorithms that might potentially lead to quantum advantage soon. The underlying ingredient is the use of informationally complete generalised measurements (IC POVMs). Our results show how hybrid variational quantum-classical algorithms using IC data allow for noise reduction, execution time reduction and efficient state preparation, paving the way to chemistry-related applications.