QuCoLiMa Topical Days

Simulating quantum many-body systems on classical computers is extremely challenging because the number of degrees of freedom scales exponentially in the system size. Already for moderate system sizes, one needs to resort to approximation methods such as Monte Carlo or Tensor Network techniques. This laboratory introduces Neural-Network Quantum States, a parametrization of the quantum states in terms of neural networks, and how can they be used to learning a steady-state of an open system with Variational Monte Carlo technique.

Goals

• Understanding the basics of variational Monte Carlo with Neural Network Quantum States
• Understanding why and when Neural Network Quantum States provide good approximations
• Running a first example code using the NetKet library, see https://www.netket.org/

Prerequisites

• Good knowledge of quantum mechanics
• Coding experience in Python
• A working installation of NetKet

Quantum computers are an exciting development and their likely first application will be in the simulation of manybody quantum systems. Some systems are available online and programming environments as well as application frameworks are become somewhat mature. Standard notions of quantum computing are taught in courses and covered in textbooks and become more or less common knowledge of quantum-minded graduate students. This session is meant to take students to the next level.

Goals:

• understanding of the implementation of optimization algorithms on gate-based and adiabatic quantum computers
• understanding of the implementation of linear algebra algorithms on gate-based quantum computers
• understanding of the variational quantum eigensolver and the variational Hamiltonian Ansatz
• practice of these algorithms in IBM QISKIT

Prerequisites:

• Understanding of the quantum gate model
• Understanding of basic quantum algorithms including
• the Grover algorithm
• Quantum phase estimation
• a working instance of IBM QISKIT at your fingertips

Mode of instruction:

Before the laboratory the students will be required to follow online lectures. The laboratory will consist of in-presence discussions. Before the class, you will receive a download link with slides and recorded audio for you to review before as well as a set of test problems on the prerequisites. In the afternoon, we will do three things:

• review and discuss the self-test problems
• discuss the material from the slides
• start working together on a new set of problems

Our trapped ion quantum computers are based on modern segmented ion traps. We will sketch architectures, the required trap technologies and fabrication methods, control electronics for quantum register reconfigurations, and recent improvements of qubit coherence and gate performance. We will present various QC applications, including variational quantum eigensolver approaches for chemistry or high energy relevant models.
The session will provide a detailed inside in the full software stack needed to operate the trapped-ion quantum computer without specific hardware knowledge and from a remote location. This includes the integration of standardized user interfaces, such as Qiskit and PennyLane, as well as the compilation stack to translate a given arbitrary quantum circuit into an optimized shuttling-based operation sequence.

Goals

• Understanding the basics of trapped-ion quantum computers
• Understanding the parts of the software stack to control a trapped-ion quantum computer, including standard user interfaces, circuit representation on different compiler layers and operation sequences to be executed on the hardware

Prerequisites

• Good knowledge of quantum mechanics
• Understanding of basic quantum algorithms